init

Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

.gemini/commands/fix-behavioral-eval.toml

@@ -0,0 +1,60 @@
+description = "Check status of nightly evals, fix failures for key models, and re-run."
+prompt = """
+You are an expert at fixing behavioral evaluations.
+
+1. **Investigate**:
+   - Use 'gh' cli to fetch the results from the latest run from the main branch: https://github.com/google-gemini/gemini-cli/actions/workflows/evals-nightly.yml.
+   - DO NOT push any changes or start any runs. The rest of your evaluation will be local.
+   - Evals are in evals/ directory and are documented by evals/README.md.
+   - The test case trajectory logs will be logged to evals/logs.
+   - You should also enable and review the verbose agent logs by setting the GEMINI_DEBUG_LOG_FILE environment variable.
+   - Identify the relevant test. Confine your investigation and validation to just this test.
+   - Proactively add logging that will aid in gathering information or validating your hypotheses.
+
+2. **Fix**:
+   - If a relevant test is failing, locate the test file and the corresponding prompt/code.
+   - It's often helpful to make an extreme, brute force change to see if you are changing the right place to make an improvement and then scope it back iteratively.
+   - Your **final** change should be **minimal and targeted**.
+   - Keep in mind the following:
+     - The prompt has multiple configurations and pieces. Take care that your changes
+       end up in the final prompt for the selected model and configuration.
+     - The prompt chosen for the eval is intentional. It's often vague or indirect
+       to see how the agent performs with ambiguous instructions. Changing it should
+       be a last resort.
+     - When changing the test prompt, carefully consider whether the prompt still tests
+       the same scenario. We don't want to lose test fidelity by making the prompts too
+       direct (i.e.: easy).
+     - Your primary mechanism for improving the agent's behavior is to make changes to
+       tool instructions, prompt.ts, and/or modules that contribute to the prompt.
+     - If prompt and description changes are unsuccessful, use logs and debugging to
+       confirm that everything is working as expected.
+    - If unable to fix the test, you can make recommendations for architecture changes
+      that might help stablize the test. Be sure to THINK DEEPLY if offering architecture guidance.
+      Some facts that might help with this are:
+      - Agents may be composed of one or more agent loops.
+      - AgentLoop == 'context + toolset + prompt'. Subagents are one type of agent loop.
+      - Agent loops perform better when:
+        - They have direct, unambiguous, and non-contradictory prompts.
+        - They have fewer irrelevant tools.
+        - They have fewer goals or steps to perform.
+        - They have less low value or irrelevant context.
+      - You may suggest compositions of existing primitives, like subagents, or
+        propose a new one.
+      - These recommendations should be high confidence and should be grounded
+        in observed deficient behaviors rather than just parroting the facts above.
+        Investigate as needed to ground your recommendations.
+
+3. **Verify**:
+   - Run just that one test if needed to validate that it is fixed. Be sure to run vitest in non-interactive mode.
+   - Running the tests can take a long time, so consider whether you can diagnose via other means or log diagnostics before committing the time. You must minimize the number of test runs needed to diagnose the failure.
+   - After the test completes, check whether it seems to have improved.
+   - You will need to run the test 3 times for Gemini 3.0, Gemini 3 flash, and Gemini 2.5 pro to ensure that it is truly stable. Run these runs in parallel, using scripts if needed.
+   - Some flakiness is expected; if it looks like a transient issue or the test is inherently unstable but passes 2/3 times, you might decide it cannot be improved.
+
+4. **Report**:
+   - Provide a summary of the test success rate for each of the tested models.
+   - Success rate is calculated based on 3 runs per model (e.g., 3/3 = 100%).
+   - If you couldn't fix it due to persistent flakiness, explain why.
+
+{{args}}
+"""

.gemini/commands/introspect.toml

@@ -0,0 +1,16 @@
+description = "Analyze the influence of system instructions on a specific action."
+prompt = """
+# Introspection Task
+
+Take a step back and analyze your own system instructions and internal logic.
+The user is curious about the reasoning behind a specific action or decision you've made.
+
+**Specific point of interest:** {{args}}
+
+Please provide a detailed breakdown of:
+1.  Which parts of your system instructions (global, workspace-specific, or provided via GEMINI.md) influenced this behavior?
+2.  What was your internal thought process leading up to this action?
+3.  Are there any ambiguities or conflicting instructions that played a role?
+
+Your goal is to provide transparency into your underlying logic so the user can potentially improve the instructions in the future.
+"""

.gemini/settings.json

@@ -0,0 +1,5 @@
+{
+  "experimental": {
+    "skills": true
+  }
+}

.gemini/skills/code-reviewer/SKILL.md

@@ -0,0 +1,65 @@
+---
+name: code-reviewer
+description:
+  Use this skill to review code. It supports both local changes (staged or working tree)
+  and remote Pull Requests (by ID or URL). It focuses on correctness, maintainability,
+  and adherence to project standards.
+---
+
+# Code Reviewer
+
+This skill guides the agent in conducting professional and thorough code reviews for both local development and remote Pull Requests.
+
+## Workflow
+
+### 1. Determine Review Target
+*   **Remote PR**: If the user provides a PR number or URL (e.g., "Review PR #123"), target that remote PR.
+*   **Local Changes**: If no specific PR is mentioned, or if the user asks to "review my changes", target the current local file system states (staged and unstaged changes).
+
+### 2. Preparation
+
+#### For Remote PRs:
+1.  **Checkout**: Use the GitHub CLI to checkout the PR.
+    ```bash
+    gh pr checkout <PR_NUMBER>
+    ```
+2.  **Preflight**: Execute the project's standard verification suite to catch automated failures early.
+    ```bash
+    npm run preflight
+    ```
+3.  **Context**: Read the PR description and any existing comments to understand the goal and history.
+
+#### For Local Changes:
+1.  **Identify Changes**:
+    *   Check status: `git status`
+    *   Read diffs: `git diff` (working tree) and/or `git diff --staged` (staged).
+2.  **Preflight (Optional)**: If the changes are substantial, ask the user if they want to run `npm run preflight` before reviewing.
+
+### 3. In-Depth Analysis
+Analyze the code changes based on the following pillars:
+
+*   **Correctness**: Does the code achieve its stated purpose without bugs or logical errors?
+*   **Maintainability**: Is the code clean, well-structured, and easy to understand and modify in the future? Consider factors like code clarity, modularity, and adherence to established design patterns.
+*   **Readability**: Is the code well-commented (where necessary) and consistently formatted according to our project's coding style guidelines?
+*   **Efficiency**: Are there any obvious performance bottlenecks or resource inefficiencies introduced by the changes?
+*   **Security**: Are there any potential security vulnerabilities or insecure coding practices?
+*   **Edge Cases and Error Handling**: Does the code appropriately handle edge cases and potential errors?
+*   **Testability**: Is the new or modified code adequately covered by tests (even if preflight checks pass)? Suggest additional test cases that would improve coverage or robustness.
+
+### 4. Provide Feedback
+
+#### Structure
+*   **Summary**: A high-level overview of the review.
+*   **Findings**:
+    *   **Critical**: Bugs, security issues, or breaking changes.
+    *   **Improvements**: Suggestions for better code quality or performance.
+    *   **Nitpicks**: Formatting or minor style issues (optional).
+*   **Conclusion**: Clear recommendation (Approved / Request Changes).
+
+#### Tone
+*   Be constructive, professional, and friendly.
+*   Explain *why* a change is requested.
+*   For approvals, acknowledge the specific value of the contribution.
+
+### 5. Cleanup (Remote PRs only)
+*   After the review, ask the user if they want to switch back to the default branch (e.g., `main` or `master`).

.gemini/skills/docs-writer/SKILL.md

@@ -0,0 +1,63 @@
+---
+name: docs-writer
+description:
+  Use this skill when asked to write documentation (`/docs` directory)
+  for Gemini CLI.
+---
+
+# `docs-writer` skill instructions
+
+As an expert technical writer for the Gemini CLI project, your goal is to
+produce documentation that is accurate, clear, and consistent with the project's
+standards. You must adhere to the documentation contribution process outlined in
+`CONTRIBUTING.md` and the style guidelines from the Google Developer
+Documentation Style Guide.
+
+## Step 1: Understand the goal and create a plan
+
+1.  **Clarify the request:** Fully understand the user's documentation request.
+    Identify the core feature, command, or concept that needs to be documented.
+2.  **Ask questions:** If the request is ambiguous or lacks detail, ask
+    clarifying questions. Don't invent or assume. It's better to ask than to
+    write incorrect documentation.
+3.  **Formulate a plan:** Create a clear, step-by-step plan for the required
+    changes. If requested or necessary, store this plan in a temporary file or
+    a file identified by the user.
+
+## Step 2: Investigate and gather information
+
+1.  **Read the code:** Thoroughly examine the relevant codebase, primarily within
+    the `packages/` directory, to ensure your writing is backed by the
+    implementation.
+2.  **Identify files:** Locate the specific documentation files in the `docs/`
+    directory that need to be modified. Always read the latest
+    version of a file before you begin to edit it.
+3.  **Check for connections:** Consider related documentation. If you add a new
+    page, check if `docs/sidebar.json` needs to be updated. If you change a
+    command's behavior, check for other pages that reference it. Make sure links
+    in these pages are up to date.
+
+## Step 3: Draft the documentation
+
+1.  **Follow the style guide:**
+    - Text must be wrapped at 80 characters. Exceptions are long links or
+      tables, unless otherwise stated by the user.
+    - Use sentence case for headings, titles, and bolded text.
+    - Address the reader as "you".
+    - Use contractions to keep the tone more casual.
+    - Use simple, direct, and active language and the present tense.
+    - Keep paragraphs short and focused.
+    - Always refer to Gemini CLI as `Gemini CLI`, never `the Gemini CLI`.
+2.  **Use `replace` and `write_file`:** Use the file system tools to apply your
+    planned changes precisely. For small edits, `replace` is preferred. For new
+    files or large rewrites, `write_file` is more appropriate.
+
+## Step 4: Verify and finalize
+
+1.  **Review your work:** After making changes, re-read the files to ensure the
+    documentation is well-formatted, content is correct and based on existing
+    code, and that all new links are valid.
+2.  **Offer to run npm format:** Once all changes are complete and the user
+    confirms they have no more requests, offer to run the project's formatting
+    script to ensure consistency. Propose the following command:
+    `npm run format`

.gemini/skills/pr-creator/SKILL.md

@@ -0,0 +1,73 @@
+---
+name: pr-creator
+description:
+  Use this skill when asked to create a pull request (PR). It ensures all PRs
+  follow the repository's established templates and standards.
+---
+
+# Pull Request Creator
+
+This skill guides the creation of high-quality Pull Requests that adhere to the
+repository's standards.
+
+## Workflow
+
+Follow these steps to create a Pull Request:
+
+1.  **Branch Management**: Check the current branch to avoid working directly
+    on `main`.
+    - Run `git branch --show-current`.
+    - If the current branch is `main`, create and switch to a new descriptive
+      branch:
+      ```bash
+      git checkout -b <new-branch-name>
+      ```
+
+2.  **Locate Template**: Search for a pull request template in the repository.
+    - Check `.github/pull_request_template.md`
+    - Check `.github/PULL_REQUEST_TEMPLATE.md`
+    - If multiple templates exist (e.g., in `.github/PULL_REQUEST_TEMPLATE/`),
+      ask the user which one to use or select the most appropriate one based on
+      the context (e.g., `bug_fix.md` vs `feature.md`).
+
+3.  **Read Template**: Read the content of the identified template file.
+
+4.  **Draft Description**: Create a PR description that strictly follows the
+    template's structure.
+    - **Headings**: Keep all headings from the template.
+    - **Checklists**: Review each item. Mark with `[x]` if completed. If an item
+      is not applicable, leave it unchecked or mark as `[ ]` (depending on the
+      template's instructions) or remove it if the template allows flexibility
+      (but prefer keeping it unchecked for transparency).
+    - **Content**: Fill in the sections with clear, concise summaries of your
+      changes.
+    - **Related Issues**: Link any issues fixed or related to this PR (e.g.,
+      "Fixes #123").
+
+5.  **Preflight Check**: Before creating the PR, run the workspace preflight
+    script to ensure all build, lint, and test checks pass.
+    ```bash
+    npm run preflight
+    ```
+    If any checks fail, address the issues before proceeding to create the PR.
+
+6.  **Create PR**: Use the `gh` CLI to create the PR. To avoid shell escaping
+    issues with multi-line Markdown, write the description to a temporary file
+    first.
+    ```bash
+    # 1. Write the drafted description to a temporary file
+    # 2. Create the PR using the --body-file flag
+    gh pr create --title "type(scope): succinct description" --body-file <temp_file_path>
+    # 3. Remove the temporary file
+    rm <temp_file_path>
+    ```
+    - **Title**: Ensure the title follows the
+      [Conventional Commits](https://www.conventionalcommits.org/) format if the
+      repository uses it (e.g., `feat(ui): add new button`,
+      `fix(core): resolve crash`).
+
+## Principles
+
+- **Compliance**: Never ignore the PR template. It exists for a reason.
+- **Completeness**: Fill out all relevant sections.
+- **Accuracy**: Don't check boxes for tasks you haven't done.

.github/ISSUE_TEMPLATE/bug_report.yml

@@ -1,7 +1,5 @@
 name: 'Bug Report'
 description: 'Report a bug to help us improve Gemini CLI'
-labels:
-  - 'status/need-triage'
 body:
   - type: 'markdown'
     attributes:

.github/dependabot.yml

@@ -1,35 +1,33 @@
-# See https://docs.github.com/en/code-security/dependabot/dependabot-version-updates/configuration-options-for-the-dependabot.yml-file
 version: 2
 updates:
   - package-ecosystem: 'npm'
     directory: '/'
     schedule:
-      interval: 'daily'
-    target-branch: 'main'
-    commit-message:
-      prefix: 'chore(deps)'
-      include: 'scope'
+      interval: 'weekly'
+      day: 'monday'
+    open-pull-requests-limit: 10
     reviewers:
-      - 'google-gemini/gemini-cli-askmode-approvers'
+      - 'joshualitt'
     groups:
-      # Group all non-major updates together.
-      # This is to reduce the number of PRs that need to be reviewed.
-      # Major updates will still be created as separate PRs.
-      npm-minor-patch:
-        applies-to: 'version-updates'
+      npm-dependencies:
+        patterns:
+          - '*'
         update-types:
           - 'minor'
           - 'patch'
-    open-pull-requests-limit: 0
 
   - package-ecosystem: 'github-actions'
     directory: '/'
     schedule:
-      interval: 'daily'
-    target-branch: 'main'
-    commit-message:
-      prefix: 'chore(deps)'
-      include: 'scope'
+      interval: 'weekly'
+      day: 'monday'
+    open-pull-requests-limit: 10
     reviewers:
-      - 'google-gemini/gemini-cli-askmode-approvers'
-    open-pull-requests-limit: 0
+      - 'joshualitt'
+    groups:
+      actions-dependencies:
+        patterns:
+          - '*'
+        update-types:
+          - 'minor'
+          - 'patch'

.github/scripts/backfill-need-triage.cjs

@@ -0,0 +1,138 @@
+/* eslint-disable */
+/* global require, console, process */
+
+/**
+ * Script to backfill the 'status/need-triage' label to all open issues
+ * that are NOT currently labeled with '🔒 maintainer only' or 'help wanted'.
+ */
+
+const { execFileSync } = require('child_process');
+
+const isDryRun = process.argv.includes('--dry-run');
+const REPO = 'google-gemini/gemini-cli';
+
+/**
+ * Executes a GitHub CLI command safely using an argument array to prevent command injection.
+ * @param {string[]} args
+ * @returns {string|null}
+ */
+function runGh(args) {
+  try {
+    // Using execFileSync with an array of arguments is safe as it doesn't use a shell.
+    // We set a large maxBuffer (10MB) to handle repositories with many issues.
+    return execFileSync('gh', args, {
+      encoding: 'utf8',
+      maxBuffer: 10 * 1024 * 1024,
+      stdio: ['ignore', 'pipe', 'pipe'],
+    }).trim();
+  } catch (error) {
+    const stderr = error.stderr ? ` Stderr: ${error.stderr.trim()}` : '';
+    console.error(
+      `❌ Error running gh ${args.join(' ')}: ${error.message}${stderr}`,
+    );
+    return null;
+  }
+}
+
+async function main() {
+  console.log('🔐 GitHub CLI security check...');
+  const authStatus = runGh(['auth', 'status']);
+  if (authStatus === null) {
+    console.error('❌ GitHub CLI (gh) is not installed or not authenticated.');
+    process.exit(1);
+  }
+
+  if (isDryRun) {
+    console.log('🧪 DRY RUN MODE ENABLED - No changes will be made.\n');
+  }
+
+  console.log(`🔍 Fetching and filtering open issues from ${REPO}...`);
+
+  // We use the /issues endpoint with pagination to bypass the 1000-result limit.
+  // The jq filter ensures we exclude PRs, maintainer-only, help-wanted, and existing status/need-triage.
+  const jqFilter =
+    '.[] | select(.pull_request == null) | select([.labels[].name] as $l | (any($l[]; . == "🔒 maintainer only") | not) and (any($l[]; . == "help wanted") | not) and (any($l[]; . == "status/need-triage") | not)) | {number: .number, title: .title}';
+
+  const output = runGh([
+    'api',
+    `repos/${REPO}/issues?state=open&per_page=100`,
+    '--paginate',
+    '--jq',
+    jqFilter,
+  ]);
+
+  if (output === null) {
+    process.exit(1);
+  }
+
+  const issues = output
+    .split('\n')
+    .filter((line) => line.trim())
+    .map((line) => {
+      try {
+        return JSON.parse(line);
+      } catch (_e) {
+        console.error(`⚠️ Failed to parse line: ${line}`);
+        return null;
+      }
+    })
+    .filter(Boolean);
+
+  console.log(`✅ Found ${issues.length} issues matching criteria.`);
+
+  if (issues.length === 0) {
+    console.log('✨ No issues need backfilling.');
+    return;
+  }
+
+  let successCount = 0;
+  let failCount = 0;
+
+  if (isDryRun) {
+    for (const issue of issues) {
+      console.log(
+        `[DRY RUN] Would label issue #${issue.number}: ${issue.title}`,
+      );
+    }
+    successCount = issues.length;
+  } else {
+    console.log(`🏷️  Applying labels to ${issues.length} issues...`);
+
+    for (const issue of issues) {
+      const issueNumber = String(issue.number);
+      console.log(`🏷️  Labeling issue #${issueNumber}: ${issue.title}`);
+
+      const result = runGh([
+        'issue',
+        'edit',
+        issueNumber,
+        '--add-label',
+        'status/need-triage',
+        '--repo',
+        REPO,
+      ]);
+
+      if (result !== null) {
+        successCount++;
+      } else {
+        failCount++;
+      }
+    }
+  }
+
+  console.log(`\n📊 Summary:`);
+  console.log(`   - Success: ${successCount}`);
+  console.log(`   - Failed:  ${failCount}`);
+
+  if (failCount > 0) {
+    console.error(`\n❌ Backfill completed with ${failCount} errors.`);
+    process.exit(1);
+  } else {
+    console.log(`\n🎉 ${isDryRun ? 'Dry run' : 'Backfill'} complete!`);
+  }
+}
+
+main().catch((error) => {
+  console.error('❌ Unexpected error:', error);
+  process.exit(1);
+});

.github/scripts/backfill-pr-notification.cjs

@@ -0,0 +1,190 @@
+/**
+ * @license
+ * Copyright 2026 Google LLC
+ * SPDX-License-Identifier: Apache-2.0
+ */
+
+/* eslint-disable */
+/* global require, console, process */
+
+/**
+ * Script to backfill a process change notification comment to all open PRs
+ * not created by members of the 'gemini-cli-maintainers' team.
+ *
+ * Skip PRs that are already associated with an issue.
+ */
+
+const { execFileSync } = require('child_process');
+
+const isDryRun = process.argv.includes('--dry-run');
+const REPO = 'google-gemini/gemini-cli';
+const ORG = 'google-gemini';
+const TEAM_SLUG = 'gemini-cli-maintainers';
+const DISCUSSION_URL =
+  'https://github.com/google-gemini/gemini-cli/discussions/16706';
+
+/**
+ * Executes a GitHub CLI command safely using an argument array.
+ */
+function runGh(args, options = {}) {
+  const { silent = false } = options;
+  try {
+    return execFileSync('gh', args, {
+      encoding: 'utf8',
+      maxBuffer: 10 * 1024 * 1024,
+      stdio: ['ignore', 'pipe', 'pipe'],
+    }).trim();
+  } catch (error) {
+    if (!silent) {
+      const stderr = error.stderr ? ` Stderr: ${error.stderr.trim()}` : '';
+      console.error(
+        `❌ Error running gh ${args.join(' ')}: ${error.message}${stderr}`,
+      );
+    }
+    return null;
+  }
+}
+
+/**
+ * Checks if a user is a member of the maintainers team.
+ */
+const membershipCache = new Map();
+function isMaintainer(username) {
+  if (membershipCache.has(username)) return membershipCache.get(username);
+
+  // GitHub returns 404 if user is not a member.
+  // We use silent: true to avoid logging 404s as errors.
+  const result = runGh(
+    ['api', `orgs/${ORG}/teams/${TEAM_SLUG}/memberships/${username}`],
+    { silent: true },
+  );
+
+  const isMember = result !== null;
+  membershipCache.set(username, isMember);
+  return isMember;
+}
+
+async function main() {
+  console.log('🔐 GitHub CLI security check...');
+  if (runGh(['auth', 'status']) === null) {
+    console.error('❌ GitHub CLI (gh) is not authenticated.');
+    process.exit(1);
+  }
+
+  if (isDryRun) {
+    console.log('🧪 DRY RUN MODE ENABLED\n');
+  }
+
+  console.log(`📥 Fetching open PRs from ${REPO}...`);
+  // Fetch number, author, and closingIssuesReferences to check if linked to an issue
+  const prsJson = runGh([
+    'pr',
+    'list',
+    '--repo',
+    REPO,
+    '--state',
+    'open',
+    '--limit',
+    '1000',
+    '--json',
+    'number,author,closingIssuesReferences',
+  ]);
+
+  if (prsJson === null) process.exit(1);
+  const prs = JSON.parse(prsJson);
+
+  console.log(`📊 Found ${prs.length} open PRs. Filtering...`);
+
+  let targetPrs = [];
+  for (const pr of prs) {
+    const author = pr.author.login;
+    const issueCount = pr.closingIssuesReferences
+      ? pr.closingIssuesReferences.length
+      : 0;
+
+    if (issueCount > 0) {
+      // Skip if already linked to an issue
+      continue;
+    }
+
+    if (!isMaintainer(author)) {
+      targetPrs.push(pr);
+    }
+  }
+
+  console.log(
+    `✅ Found ${targetPrs.length} PRs from non-maintainers without associated issues.`,
+  );
+
+  const commentBody =
+    "\nHi @{AUTHOR}, thank you so much for your contribution to Gemini CLI! We really appreciate the time and effort you've put into this.\n\nWe're making some updates to our contribution process to improve how we track and review changes. Please take a moment to review our recent discussion post: [Improving Our Contribution Process & Introducing New Guidelines](${DISCUSSION_URL}).\n\nKey Update: Starting **January 26, 2026**, the Gemini CLI project will require all pull requests to be associated with an existing issue. Any pull requests not linked to an issue by that date will be automatically closed.\n\nThank you for your understanding and for being a part of our community!\n  ".trim();
+
+  let successCount = 0;
+  let skipCount = 0;
+  let failCount = 0;
+
+  for (const pr of targetPrs) {
+    const prNumber = String(pr.number);
+    const author = pr.author.login;
+
+    // Check if we already commented (idempotency)
+    // We use silent: true here because view might fail if PR is deleted mid-run
+    const existingComments = runGh(
+      [
+        'pr',
+        'view',
+        prNumber,
+        '--repo',
+        REPO,
+        '--json',
+        'comments',
+        '--jq',
+        `.comments[].body | contains("${DISCUSSION_URL}")`,
+      ],
+      { silent: true },
+    );
+
+    if (existingComments && existingComments.includes('true')) {
+      console.log(
+        `⏭️  PR #${prNumber} already has the notification. Skipping.`,
+      );
+      skipCount++;
+      continue;
+    }
+
+    if (isDryRun) {
+      console.log(`[DRY RUN] Would notify @${author} on PR #${prNumber}`);
+      successCount++;
+    } else {
+      console.log(`💬 Notifying @${author} on PR #${prNumber}...`);
+      const personalizedComment = commentBody.replace('{AUTHOR}', author);
+      const result = runGh([
+        'pr',
+        'comment',
+        prNumber,
+        '--repo',
+        REPO,
+        '--body',
+        personalizedComment,
+      ]);
+
+      if (result !== null) {
+        successCount++;
+      } else {
+        failCount++;
+      }
+    }
+  }
+
+  console.log(`\n📊 Summary:`);
+  console.log(`   - Notified: ${successCount}`);
+  console.log(`   - Skipped:  ${skipCount}`);
+  console.log(`   - Failed:   ${failCount}`);
+
+  if (failCount > 0) process.exit(1);
+}
+
+main().catch((e) => {
+  console.error(e);
+  process.exit(1);
+});

.github/scripts/pr-triage.sh

@@ -1,162 +1,180 @@
 #!/usr/bin/env bash
+# @license
+# Copyright 2026 Google LLC
+# SPDX-License-Identifier: Apache-2.0
+
 set -euo pipefail
 
 # Initialize a comma-separated string to hold PR numbers that need a comment
 PRS_NEEDING_COMMENT=""
 
-# Function to process a single PR
-process_pr() {
-    if [[ -z "${GITHUB_REPOSITORY:-}" ]]; then
-        echo "‼️ Missing \$GITHUB_REPOSITORY - this must be run from GitHub Actions"
-        return 1
+# Global cache for issue labels (compatible with Bash 3.2)
+# Stores "|ISSUE_NUM:LABELS|" segments
+ISSUE_LABELS_CACHE_FLAT="|"
+
+# Function to get labels from an issue (with caching)
+get_issue_labels() {
+    local ISSUE_NUM="${1}"
+    if [[ -z "${ISSUE_NUM}" || "${ISSUE_NUM}" == "null" || "${ISSUE_NUM}" == "" ]]; then
+        return
     fi
 
-    if [[ -z "${GITHUB_OUTPUT:-}" ]]; then
-        echo "‼️ Missing \$GITHUB_OUTPUT - this must be run from GitHub Actions"
-        return 1
+    # Check cache
+    case "${ISSUE_LABELS_CACHE_FLAT}" in
+        *"|${ISSUE_NUM}:"*) 
+            local suffix="${ISSUE_LABELS_CACHE_FLAT#*|${ISSUE_NUM}:}"
+            echo "${suffix%%|*}"
+            return
+            ;; 
+        *)
+            # Cache miss, proceed to fetch
+            ;;
+    esac
+
+    echo "   📥 Fetching labels from issue #${ISSUE_NUM}" >&2
+    local gh_output
+    if ! gh_output=$(gh issue view "${ISSUE_NUM}" --repo "${GITHUB_REPOSITORY}" --json labels -q '.labels[].name' 2>/dev/null); then
+        echo "      ⚠️ Could not fetch issue #${ISSUE_NUM}" >&2
+        ISSUE_LABELS_CACHE_FLAT="${ISSUE_LABELS_CACHE_FLAT}${ISSUE_NUM}:|"
+        return
     fi
 
-    local PR_NUMBER=$1
+    local labels
+    labels=$(echo "${gh_output}" | grep -x -E '(area|priority)/.*|help wanted|🔒 maintainer only' | tr '\n' ',' | sed 's/,$//' || echo "")
+    
+    # Save to flat cache
+    ISSUE_LABELS_CACHE_FLAT="${ISSUE_LABELS_CACHE_FLAT}${ISSUE_NUM}:${labels}|"
+    echo "${labels}"
+}
+
+# Function to process a single PR with pre-fetched data
+process_pr_optimized() {
+    local PR_NUMBER="${1}"
+    local IS_DRAFT="${2}"
+    local ISSUE_NUMBER="${3}"
+    local CURRENT_LABELS="${4}" # Comma-separated labels
+
     echo "🔄 Processing PR #${PR_NUMBER}"
 
-    # Get PR details: closing issue, draft status, body and labels
-    local PR_DATA
-    if ! PR_DATA=$(gh pr view "${PR_NUMBER}" --repo "${GITHUB_REPOSITORY}" --json closingIssuesReferences,isDraft,body,labels 2>/dev/null); then
-        echo "   ⚠️ Could not fetch data for PR #${PR_NUMBER}"
-        return 0
-    fi
+    local LABELS_TO_ADD=""
+    local LABELS_TO_REMOVE=""
 
-    local ISSUE_NUMBER
-    ISSUE_NUMBER=$(echo "${PR_DATA}" | jq -r '.closingIssuesReferences[0].number // empty')
-
-    # If no closing issue found, check body for references (e.g. #123)
-    if [[ -z "${ISSUE_NUMBER}" ]]; then
-        local REFERENCED_ISSUE
-        # Search for # followed by digits, not preceded by alphanumeric chars
-        REFERENCED_ISSUE=$(echo "${PR_DATA}" | jq -r '.body // empty' | grep -oE '(^|[^a-zA-Z0-9])#[0-9]+([^a-zA-Z0-9]|$)' | head -n 1 | grep -oE '[0-9]+' || echo "")
-        if [[ -n "${REFERENCED_ISSUE}" ]]; then
-            ISSUE_NUMBER="${REFERENCED_ISSUE}"
-            echo "🔗 Found referenced issue #${ISSUE_NUMBER} in PR body"
-        fi
-    fi
-
-    local IS_DRAFT
-    IS_DRAFT=$(echo "${PR_DATA}" | jq -r '.isDraft')
-
-    if [[ -z "${ISSUE_NUMBER}" ]]; then
+    if [[ -z "${ISSUE_NUMBER}" || "${ISSUE_NUMBER}" == "null" || "${ISSUE_NUMBER}" == "" ]]; then
         if [[ "${IS_DRAFT}" == "true" ]]; then
-            echo "📝 PR #${PR_NUMBER} is a draft and has no linked issue, skipping status/need-issue label"
-            # Remove status/need-issue label if it was previously added
-            if ! gh pr edit "${PR_NUMBER}" --repo "${GITHUB_REPOSITORY}" --remove-label "status/need-issue" 2>/dev/null; then
-                echo "   status/need-issue label not present or could not be removed"
+            echo "   📝 PR #${PR_NUMBER} is a draft and has no linked issue"
+            if [[ ",${CURRENT_LABELS}," == *",status/need-issue,"* ]]; then
+                echo "      ➖ Removing status/need-issue label"
+                LABELS_TO_REMOVE="status/need-issue"
             fi
-            echo "needs_comment=false" >> "${GITHUB_OUTPUT}"
         else
-            echo "⚠️  No linked issue found for PR #${PR_NUMBER}, adding status/need-issue label"
-            if ! gh pr edit "${PR_NUMBER}" --repo "${GITHUB_REPOSITORY}" --add-label "status/need-issue" 2>/dev/null; then
-                echo "   ⚠️ Failed to add label (may already exist or have permission issues)"
+            echo "   ⚠️  No linked issue found for PR #${PR_NUMBER}"
+            if [[ ",${CURRENT_LABELS}," != *",status/need-issue,"* ]]; then
+                echo "      ➕ Adding status/need-issue label"
+                LABELS_TO_ADD="status/need-issue"
             fi
-            # Add PR number to the list
+            
             if [[ -z "${PRS_NEEDING_COMMENT}" ]]; then
                 PRS_NEEDING_COMMENT="${PR_NUMBER}"
             else
                 PRS_NEEDING_COMMENT="${PRS_NEEDING_COMMENT},${PR_NUMBER}"
             fi
-            echo "needs_comment=true" >> "${GITHUB_OUTPUT}"
         fi
     else
-        echo "🔗 Found linked issue #${ISSUE_NUMBER}"
+        echo "   🔗 Found linked issue #${ISSUE_NUMBER}"
 
-        # Remove status/need-issue label if present
-        if ! gh pr edit "${PR_NUMBER}" --repo "${GITHUB_REPOSITORY}" --remove-label "status/need-issue" 2>/dev/null; then
-            echo "   status/need-issue label not present or could not be removed"
+        if [[ ",${CURRENT_LABELS}," == *",status/need-issue,"* ]]; then
+            echo "      ➖ Removing status/need-issue label"
+            LABELS_TO_REMOVE="status/need-issue"
         fi
 
-        # Get issue labels
-        echo "📥 Fetching area and priority labels from issue #${ISSUE_NUMBER}"
-        local ISSUE_LABELS=""
-        local gh_output
-        if ! gh_output=$(gh issue view "${ISSUE_NUMBER}" --repo "${GITHUB_REPOSITORY}" --json labels -q '.labels[].name' 2>/dev/null); then
-            echo "   ⚠️ Could not fetch issue #${ISSUE_NUMBER} (may not exist or be in different repo)"
-            ISSUE_LABELS=""
-        else
-            # If grep finds no matches, it exits with 1, which pipefail would treat as an error.
-            # `|| echo ""` ensures the command succeeds with an empty string in that case.
-            ISSUE_LABELS=$(echo "${gh_output}" | grep -E "^(area|priority)/" | tr '\n' ',' | sed 's/,$//' || echo "")
-        fi
+        local ISSUE_LABELS
+        ISSUE_LABELS=$(get_issue_labels "${ISSUE_NUMBER}")
 
-        # Get PR labels from already fetched PR_DATA
-        echo "📥 Extracting labels from PR #${PR_NUMBER}"
-        local PR_LABELS=""
-        PR_LABELS=$(echo "${PR_DATA}" | jq -r '.labels[].name // empty' | tr '\n' ',' | sed 's/,$//' || echo "")
-
-        echo "   Issue labels (area/priority): ${ISSUE_LABELS}"
-        echo "   PR labels: ${PR_LABELS}"
-
-        # Convert comma-separated strings to arrays
-        local ISSUE_LABEL_ARRAY PR_LABEL_ARRAY
-        IFS=',' read -ra ISSUE_LABEL_ARRAY <<< "${ISSUE_LABELS}"
-        IFS=',' read -ra PR_LABEL_ARRAY <<< "${PR_LABELS:-}"
-
-        # Find labels to add (on issue but not on PR)
-        local LABELS_TO_ADD=""
-        for label in "${ISSUE_LABEL_ARRAY[@]}"; do
-            if [[ -n "${label}" ]] && [[ " ${PR_LABEL_ARRAY[*]:-}" != *" ${label} "* ]]; then
-                if [[ -z "${LABELS_TO_ADD}" ]]; then
-                    LABELS_TO_ADD="${label}"
-                else
-                    LABELS_TO_ADD="${LABELS_TO_ADD},${label}"
+        if [[ -n "${ISSUE_LABELS}" ]]; then
+            local IFS_OLD="${IFS}"
+            IFS=','
+            for label in ${ISSUE_LABELS}; do
+                if [[ -n "${label}" ]] && [[ ",${CURRENT_LABELS}," != *",${label},"* ]]; then
+                    if [[ -z "${LABELS_TO_ADD}" ]]; then
+                        LABELS_TO_ADD="${label}"
+                    else
+                        LABELS_TO_ADD="${LABELS_TO_ADD},${label}"
+                    fi
                 fi
-            fi
-        done
+done
+            IFS="${IFS_OLD}"
+        fi
 
-        # Apply label changes
+        if [[ -z "${LABELS_TO_ADD}" && -z "${LABELS_TO_REMOVE}" ]]; then
+            echo "   ✅ Labels already synchronized"
+        fi
+    fi
+
+    if [[ -n "${LABELS_TO_ADD}" || -n "${LABELS_TO_REMOVE}" ]]; then
+        local EDIT_CMD=("gh" "pr" "edit" "${PR_NUMBER}" "--repo" "${GITHUB_REPOSITORY}")
         if [[ -n "${LABELS_TO_ADD}" ]]; then
-            echo "➕ Adding labels: ${LABELS_TO_ADD}"
-            if ! gh pr edit "${PR_NUMBER}" --repo "${GITHUB_REPOSITORY}" --add-label "${LABELS_TO_ADD}" 2>/dev/null; then
-                echo "   ⚠️ Failed to add some labels"
-            fi
+            echo "      ➕ Syncing labels to add: ${LABELS_TO_ADD}"
+            EDIT_CMD+=("--add-label" "${LABELS_TO_ADD}")
         fi
-
-        if [[ -z "${LABELS_TO_ADD}" ]]; then
-            echo "✅ Labels already synchronized"
+        if [[ -n "${LABELS_TO_REMOVE}" ]]; then
+            echo "      ➖ Syncing labels to remove: ${LABELS_TO_REMOVE}"
+            EDIT_CMD+=("--remove-label" "${LABELS_TO_REMOVE}")
         fi
-        echo "needs_comment=false" >> "${GITHUB_OUTPUT}"
+        
+        ("${EDIT_CMD[@]}" || true)
     fi
 }
 
-# If PR_NUMBER is set, process only that PR
-if [[ -n "${PR_NUMBER:-}" ]]; then
-    if ! process_pr "${PR_NUMBER}"; then
-        echo "❌ Failed to process PR #${PR_NUMBER}"
-        exit 1
-    fi
-else
-    # Otherwise, get all open PRs and process them
-    # The script logic will determine which ones need issue linking or label sync
-    echo "📥 Getting all open pull requests..."
-    if ! PR_NUMBERS=$(gh pr list --repo "${GITHUB_REPOSITORY}" --state open --limit 1000 --json number -q '.[].number' 2>/dev/null); then
-        echo "❌ Failed to fetch PR list"
-        exit 1
-    fi
-
-    if [[ -z "${PR_NUMBERS}" ]]; then
-        echo "✅ No open PRs found"
-    else
-        # Count the number of PRs
-        PR_COUNT=$(echo "${PR_NUMBERS}" | wc -w | tr -d ' ')
-        echo "📊 Found ${PR_COUNT} open PRs to process"
-
-        for pr_number in ${PR_NUMBERS}; do
-            if ! process_pr "${pr_number}"; then
-                echo "⚠️ Failed to process PR #${pr_number}, continuing with next PR..."
-                continue
-            fi
-        done
-    fi
+if [[ -z "${GITHUB_REPOSITORY:-}" ]]; then
+    echo "‼️ Missing \$GITHUB_REPOSITORY - this must be run from GitHub Actions"
+    exit 1
+fi
+
+if [[ -z "${GITHUB_OUTPUT:-}" ]]; then
+    echo "‼️ Missing \$GITHUB_OUTPUT - this must be run from GitHub Actions"
+    exit 1
+fi
+
+JQ_EXTRACT_FIELDS='{
+    number: .number,
+    isDraft: .isDraft,
+    issue: (.closingIssuesReferences[0].number // (.body // "" | capture("(^|[^a-zA-Z0-9])#(?<num>[0-9]+)([^a-zA-Z0-9]|$)")? | .num) // "null"),
+    labels: [.labels[].name] | join(",")
+}'
+
+JQ_TSV_FORMAT='"\((.number | tostring))\t\(.isDraft)\t\((.issue // null) | tostring)\t\(.labels)"'
+
+if [[ -n "${PR_NUMBER:-}" ]]; then
+    echo "🔄 Processing single PR #${PR_NUMBER}"
+    PR_DATA=$(gh pr view "${PR_NUMBER}" --repo "${GITHUB_REPOSITORY}" --json number,closingIssuesReferences,isDraft,body,labels 2>/dev/null) || {
+        echo "❌ Failed to fetch data for PR #${PR_NUMBER}"
+        exit 1
+    }
+    
+    line=$(echo "${PR_DATA}" | jq -r "${JQ_EXTRACT_FIELDS} | ${JQ_TSV_FORMAT}")
+    IFS=$'\t' read -r pr_num is_draft issue_num current_labels <<< "${line}"
+    process_pr_optimized "${pr_num}" "${is_draft}" "${issue_num}" "${current_labels}"
+else
+    echo "📥 Getting all open pull requests..."
+    PR_DATA_ALL=$(gh pr list --repo "${GITHUB_REPOSITORY}" --state open --limit 1000 --json number,closingIssuesReferences,isDraft,body,labels 2>/dev/null) || {
+        echo "❌ Failed to fetch PR list"
+        exit 1
+    }
+
+    PR_COUNT=$(echo "${PR_DATA_ALL}" | jq '. | length')
+    echo "📊 Found ${PR_COUNT} open PRs to process"
+
+    # Use a temporary file to avoid masking exit codes in process substitution
+    tmp_file=$(mktemp)
+    echo "${PR_DATA_ALL}" | jq -r ".[] | ${JQ_EXTRACT_FIELDS} | ${JQ_TSV_FORMAT}" > "${tmp_file}"
+    while read -r line; do
+        [[ -z "${line}" ]] && continue
+        IFS=$'\t' read -r pr_num is_draft issue_num current_labels <<< "${line}"
+        process_pr_optimized "${pr_num}" "${is_draft}" "${issue_num}" "${current_labels}"
+    done < "${tmp_file}"
+    rm -f "${tmp_file}"
 fi
 
-# Ensure output is always set, even if empty
 if [[ -z "${PRS_NEEDING_COMMENT}" ]]; then
     echo "prs_needing_comment=[]" >> "${GITHUB_OUTPUT}"
 else

.github/scripts/sync-maintainer-labels.cjs

@@ -0,0 +1,355 @@
+/* eslint-disable @typescript-eslint/no-require-imports */
+/* global process, console, require */
+const { Octokit } = require('@octokit/rest');
+
+/**
+ * Sync Maintainer Labels (Recursive with strict parent-child relationship detection)
+ * - Uses Native Sub-issues.
+ * - Uses Markdown Task Lists (- [ ] #123).
+ * - Filters for OPEN issues only.
+ * - Skips DUPLICATES.
+ * - Skips Pull Requests.
+ * - ONLY labels issues in the PUBLIC (gemini-cli) repo.
+ */
+
+const REPO_OWNER = 'google-gemini';
+const PUBLIC_REPO = 'gemini-cli';
+const PRIVATE_REPO = 'maintainers-gemini-cli';
+const ALLOWED_REPOS = [PUBLIC_REPO, PRIVATE_REPO];
+
+const ROOT_ISSUES = [
+  { owner: REPO_OWNER, repo: PUBLIC_REPO, number: 15374 },
+  { owner: REPO_OWNER, repo: PUBLIC_REPO, number: 15456 },
+  { owner: REPO_OWNER, repo: PUBLIC_REPO, number: 15324 },
+];
+
+const TARGET_LABEL = '🔒 maintainer only';
+const isDryRun =
+  process.argv.includes('--dry-run') || process.env.DRY_RUN === 'true';
+
+const octokit = new Octokit({
+  auth: process.env.GITHUB_TOKEN,
+});
+
+/**
+ * Extracts child issue references from markdown Task Lists ONLY.
+ * e.g. - [ ] #123 or - [x] google-gemini/gemini-cli#123
+ */
+function extractTaskListLinks(text, contextOwner, contextRepo) {
+  if (!text) return [];
+  const childIssues = new Map();
+
+  const add = (owner, repo, number) => {
+    if (ALLOWED_REPOS.includes(repo)) {
+      const key = `${owner}/${repo}#${number}`;
+      childIssues.set(key, { owner, repo, number: parseInt(number, 10) });
+    }
+  };
+
+  // 1. Full URLs in task lists
+  const urlRegex =
+    /-\s+\[[ x]\].*https:\/\/github\.com\/([a-zA-Z0-9._-]+)\/([a-zA-Z0-9._-]+)\/issues\/(\d+)\b/g;
+  let match;
+  while ((match = urlRegex.exec(text)) !== null) {
+    add(match[1], match[2], match[3]);
+  }
+
+  // 2. Cross-repo refs in task lists: owner/repo#123
+  const crossRepoRegex =
+    /-\s+\[[ x]\].*([a-zA-Z0-9._-]+)\/([a-zA-Z0-9._-]+)#(\d+)\b/g;
+  while ((match = crossRepoRegex.exec(text)) !== null) {
+    add(match[1], match[2], match[3]);
+  }
+
+  // 3. Short refs in task lists: #123
+  const shortRefRegex = /-\s+\[[ x]\].*#(\d+)\b/g;
+  while ((match = shortRefRegex.exec(text)) !== null) {
+    add(contextOwner, contextRepo, match[1]);
+  }
+
+  return Array.from(childIssues.values());
+}
+
+/**
+ * Fetches issue data via GraphQL with full pagination for sub-issues, comments, and labels.
+ */
+async function fetchIssueData(owner, repo, number) {
+  const query = `
+    query($owner:String!, $repo:String!, $number:Int!) {
+      repository(owner:$owner, name:$repo) {
+        issue(number:$number) {
+          state
+          title
+          body
+          labels(first: 100) {
+            nodes { name }
+            pageInfo { hasNextPage endCursor }
+          }
+          subIssues(first: 100) {
+            nodes {
+              number
+              repository {
+                name
+                owner { login }
+              }
+            }
+            pageInfo { hasNextPage endCursor }
+          }
+          comments(first: 100) {
+            nodes {
+              body
+            }
+          }
+        }
+      }
+    }
+  `;
+
+  try {
+    const response = await octokit.graphql(query, { owner, repo, number });
+    const data = response.repository.issue;
+    if (!data) return null;
+
+    const issue = {
+      state: data.state,
+      title: data.title,
+      body: data.body || '',
+      labels: data.labels.nodes.map((n) => n.name),
+      subIssues: [...data.subIssues.nodes],
+      comments: data.comments.nodes.map((n) => n.body),
+    };
+
+    // Paginate subIssues if there are more than 100
+    if (data.subIssues.pageInfo.hasNextPage) {
+      const moreSubIssues = await paginateConnection(
+        owner,
+        repo,
+        number,
+        'subIssues',
+        'number repository { name owner { login } }',
+        data.subIssues.pageInfo.endCursor,
+      );
+      issue.subIssues.push(...moreSubIssues);
+    }
+
+    // Paginate labels if there are more than 100 (unlikely but for completeness)
+    if (data.labels.pageInfo.hasNextPage) {
+      const moreLabels = await paginateConnection(
+        owner,
+        repo,
+        number,
+        'labels',
+        'name',
+        data.labels.pageInfo.endCursor,
+        (n) => n.name,
+      );
+      issue.labels.push(...moreLabels);
+    }
+
+    // Note: Comments are handled via Task Lists in body + first 100 comments.
+    // If an issue has > 100 comments with task lists, we'd need to paginate those too.
+    // Given the 1,100+ issue discovery count, 100 comments is usually sufficient,
+    // but we can add it for absolute completeness.
+    // (Skipping for now to avoid excessive API churn unless clearly needed).
+
+    return issue;
+  } catch (error) {
+    if (error.errors && error.errors.some((e) => e.type === 'NOT_FOUND')) {
+      return null;
+    }
+    throw error;
+  }
+}
+
+/**
+ * Helper to paginate any GraphQL connection.
+ */
+async function paginateConnection(
+  owner,
+  repo,
+  number,
+  connectionName,
+  nodeFields,
+  initialCursor,
+  transformNode = (n) => n,
+) {
+  let additionalNodes = [];
+  let hasNext = true;
+  let cursor = initialCursor;
+
+  while (hasNext) {
+    const query = `
+      query($owner:String!, $repo:String!, $number:Int!, $cursor:String) {
+        repository(owner:$owner, name:$repo) {
+          issue(number:$number) {
+            ${connectionName}(first: 100, after: $cursor) {
+              nodes { ${nodeFields} }
+              pageInfo { hasNextPage endCursor }
+            }
+          }
+        }
+      }
+    `;
+    const response = await octokit.graphql(query, {
+      owner,
+      repo,
+      number,
+      cursor,
+    });
+    const connection = response.repository.issue[connectionName];
+    additionalNodes.push(...connection.nodes.map(transformNode));
+    hasNext = connection.pageInfo.hasNextPage;
+    cursor = connection.pageInfo.endCursor;
+  }
+  return additionalNodes;
+}
+
+/**
+ * Validates if an issue should be processed (Open, not a duplicate, not a PR)
+ */
+function shouldProcess(issueData) {
+  if (!issueData) return false;
+
+  if (issueData.state !== 'OPEN') return false;
+
+  const labels = issueData.labels.map((l) => l.toLowerCase());
+  if (labels.includes('duplicate') || labels.includes('kind/duplicate')) {
+    return false;
+  }
+
+  return true;
+}
+
+async function getAllDescendants(roots) {
+  const allDescendants = new Map();
+  const visited = new Set();
+  const queue = [...roots];
+
+  for (const root of roots) {
+    visited.add(`${root.owner}/${root.repo}#${root.number}`);
+  }
+
+  console.log(`Starting discovery from ${roots.length} roots...`);
+
+  while (queue.length > 0) {
+    const current = queue.shift();
+    const currentKey = `${current.owner}/${current.repo}#${current.number}`;
+
+    try {
+      const issueData = await fetchIssueData(
+        current.owner,
+        current.repo,
+        current.number,
+      );
+
+      if (!shouldProcess(issueData)) {
+        continue;
+      }
+
+      // ONLY add to labeling list if it's in the PUBLIC repository
+      if (current.repo === PUBLIC_REPO) {
+        // Don't label the roots themselves
+        if (
+          !ROOT_ISSUES.some(
+            (r) => r.number === current.number && r.repo === current.repo,
+          )
+        ) {
+          allDescendants.set(currentKey, {
+            ...current,
+            title: issueData.title,
+            labels: issueData.labels,
+          });
+        }
+      }
+
+      const children = new Map();
+
+      // 1. Process Native Sub-issues
+      if (issueData.subIssues) {
+        for (const node of issueData.subIssues) {
+          const childOwner = node.repository.owner.login;
+          const childRepo = node.repository.name;
+          const childNumber = node.number;
+          const key = `${childOwner}/${childRepo}#${childNumber}`;
+          children.set(key, {
+            owner: childOwner,
+            repo: childRepo,
+            number: childNumber,
+          });
+        }
+      }
+
+      // 2. Process Markdown Task Lists in Body and Comments
+      let combinedText = issueData.body || '';
+      if (issueData.comments) {
+        for (const commentBody of issueData.comments) {
+          combinedText += '\n' + (commentBody || '');
+        }
+      }
+
+      const taskListLinks = extractTaskListLinks(
+        combinedText,
+        current.owner,
+        current.repo,
+      );
+      for (const link of taskListLinks) {
+        const key = `${link.owner}/${link.repo}#${link.number}`;
+        children.set(key, link);
+      }
+
+      // Queue children (regardless of which repo they are in, for recursion)
+      for (const [key, child] of children) {
+        if (!visited.has(key)) {
+          visited.add(key);
+          queue.push(child);
+        }
+      }
+    } catch (error) {
+      console.error(`Error processing ${currentKey}: ${error.message}`);
+    }
+  }
+
+  return Array.from(allDescendants.values());
+}
+
+async function run() {
+  if (isDryRun) {
+    console.log('=== DRY RUN MODE: No labels will be applied ===');
+  }
+
+  const descendants = await getAllDescendants(ROOT_ISSUES);
+  console.log(
+    `\nFound ${descendants.length} total unique open descendant issues in ${PUBLIC_REPO}.`,
+  );
+
+  for (const issueInfo of descendants) {
+    const issueKey = `${issueInfo.owner}/${issueInfo.repo}#${issueInfo.number}`;
+    try {
+      // Data is already available from the discovery phase
+      const hasLabel = issueInfo.labels.some((l) => l === TARGET_LABEL);
+
+      if (!hasLabel) {
+        if (isDryRun) {
+          console.log(
+            `[DRY RUN] Would label ${issueKey}: "${issueInfo.title}"`,
+          );
+        } else {
+          console.log(`Labeling ${issueKey}: "${issueInfo.title}"...`);
+          await octokit.rest.issues.addLabels({
+            owner: issueInfo.owner,
+            repo: issueInfo.repo,
+            issue_number: issueInfo.number,
+            labels: [TARGET_LABEL],
+          });
+        }
+      }
+    } catch (error) {
+      console.error(`Error processing label for ${issueKey}: ${error.message}`);
+    }
+  }
+}
+
+run().catch((error) => {
+  console.error(error);
+  process.exit(1);
+});

.github/workflows/chained_e2e.yml

@@ -277,6 +277,37 @@ jobs:
         shell: 'pwsh'
         run: 'npm run test:integration:sandbox:none'
 
+  evals:
+    name: 'Evals (ALWAYS_PASSING)'
+    needs:
+      - 'merge_queue_skipper'
+      - 'parse_run_context'
+    runs-on: 'gemini-cli-ubuntu-16-core'
+    if: |
+      always() && (needs.merge_queue_skipper.result !='success' || needs.merge_queue_skipper.outputs.skip != 'true')
+    steps:
+      - name: 'Checkout'
+        uses: 'actions/checkout@08eba0b27e820071cde6df949e0beb9ba4906955' # ratchet:actions/checkout@v5
+        with:
+          ref: '${{ needs.parse_run_context.outputs.sha }}'
+          repository: '${{ needs.parse_run_context.outputs.repository }}'
+
+      - name: 'Set up Node.js 20.x'
+        uses: 'actions/setup-node@49933ea5288caeca8642d1e84afbd3f7d6820020' # ratchet:actions-node@v4
+        with:
+          node-version: '20.x'
+
+      - name: 'Install dependencies'
+        run: 'npm ci'
+
+      - name: 'Build project'
+        run: 'npm run build'
+
+      - name: 'Run Evals (Required to pass)'
+        env:
+          GEMINI_API_KEY: '${{ secrets.GEMINI_API_KEY }}'
+        run: 'npm run test:always_passing_evals'
+
   e2e:
     name: 'E2E'
     if: |
@@ -284,13 +315,15 @@ jobs:
     needs:
       - 'e2e_linux'
       - 'e2e_mac'
+      - 'evals'
       - 'merge_queue_skipper'
     runs-on: 'gemini-cli-ubuntu-16-core'
     steps:
       - name: 'Check E2E test results'
         run: |
           if [[ ${{ needs.e2e_linux.result }} != 'success' || \
-               ${{ needs.e2e_mac.result }} != 'success' ]]; then
+               ${{ needs.e2e_mac.result }} != 'success' || \
+               ${{ needs.evals.result }} != 'success' ]]; then
             echo "One or more E2E jobs failed."
             exit 1
           fi

.github/workflows/evals-nightly.yml

@@ -0,0 +1,101 @@
+name: 'Evals: Nightly'
+
+on:
+  schedule:
+    - cron: '0 1 * * *' # Runs at 1 AM every day
+  workflow_dispatch:
+    inputs:
+      run_all:
+        description: 'Run all evaluations (including usually passing)'
+        type: 'boolean'
+        default: true
+      test_name_pattern:
+        description: 'Test name pattern or file name'
+        required: false
+        type: 'string'
+
+permissions:
+  contents: 'read'
+  checks: 'write'
+  actions: 'read'
+
+jobs:
+  evals:
+    name: 'Evals (USUALLY_PASSING) nightly run'
+    runs-on: 'gemini-cli-ubuntu-16-core'
+    strategy:
+      fail-fast: false
+      matrix:
+        model:
+          - 'gemini-3-pro-preview'
+          - 'gemini-3-flash-preview'
+          - 'gemini-2.5-pro'
+          - 'gemini-2.5-flash'
+          - 'gemini-2.5-flash-lite'
+        run_attempt: [1, 2, 3]
+    steps:
+      - name: 'Checkout'
+        uses: 'actions/checkout@08c6903cd8c0fde910a37f88322edcfb5dd907a8' # ratchet:actions/checkout@v5
+
+      - name: 'Set up Node.js'
+        uses: 'actions/setup-node@49933ea5288caeca8642d1e84afbd3f7d6820020' # ratchet:actions/setup-node@v4
+        with:
+          node-version-file: '.nvmrc'
+          cache: 'npm'
+
+      - name: 'Install dependencies'
+        run: 'npm ci'
+
+      - name: 'Build project'
+        run: 'npm run build'
+
+      - name: 'Create logs directory'
+        run: 'mkdir -p evals/logs'
+
+      - name: 'Run Evals'
+        continue-on-error: true
+        env:
+          GEMINI_API_KEY: '${{ secrets.GEMINI_API_KEY }}'
+          GEMINI_MODEL: '${{ matrix.model }}'
+          RUN_EVALS: "${{ github.event.inputs.run_all != 'false' }}"
+          TEST_NAME_PATTERN: '${{ github.event.inputs.test_name_pattern }}'
+        run: |
+          CMD="npm run test:all_evals"
+          PATTERN="${{ env.TEST_NAME_PATTERN }}"
+
+          if [[ -n "$PATTERN" ]]; then
+            if [[ "$PATTERN" == *.ts || "$PATTERN" == *.js || "$PATTERN" == */* ]]; then
+              $CMD -- "$PATTERN"
+            else
+              $CMD -- -t "$PATTERN"
+            fi
+          else
+            $CMD
+          fi
+
+      - name: 'Upload Logs'
+        if: 'always()'
+        uses: 'actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02' # ratchet:actions/upload-artifact@v4
+        with:
+          name: 'eval-logs-${{ matrix.model }}-${{ matrix.run_attempt }}'
+          path: 'evals/logs'
+          retention-days: 7
+
+  aggregate-results:
+    name: 'Aggregate Results'
+    needs: ['evals']
+    if: 'always()'
+    runs-on: 'gemini-cli-ubuntu-16-core'
+    steps:
+      - name: 'Checkout'
+        uses: 'actions/checkout@08c6903cd8c0fde910a37f88322edcfb5dd907a8' # ratchet:actions/checkout@v5
+
+      - name: 'Download Logs'
+        uses: 'actions/download-artifact@cc203385981b70ca67e1cc392babf9cc229d5806' # ratchet:actions/download-artifact@v4
+        with:
+          path: 'artifacts'
+
+      - name: 'Generate Summary'
+        env:
+          GH_TOKEN: '${{ secrets.GITHUB_TOKEN }}'
+        run: 'node scripts/aggregate_evals.js artifacts >> "$GITHUB_STEP_SUMMARY"'

.github/workflows/gemini-automated-issue-triage.yml

@@ -14,9 +14,15 @@ on:
         description: 'issue number to triage'
         required: true
         type: 'number'
+  workflow_call:
+    inputs:
+      issue_number:
+        description: 'issue number to triage'
+        required: false
+        type: 'string'
 
 concurrency:
-  group: '${{ github.workflow }}-${{ github.event.issue.number || github.event.inputs.issue_number }}'
+  group: '${{ github.workflow }}-${{ github.event.issue.number || github.event.inputs.issue_number || inputs.issue_number }}'
   cancel-in-progress: true
 
 defaults:
@@ -34,7 +40,7 @@ permissions:
 jobs:
   triage-issue:
     if: |-
-      github.repository == 'google-gemini/gemini-cli' &&
+      (github.repository == 'google-gemini/gemini-cli' || github.repository == 'google-gemini/maintainers-gemini-cli') &&
       (
         github.event_name == 'workflow_dispatch' ||
         (
@@ -57,10 +63,11 @@ jobs:
         with:
           github-token: '${{ secrets.GITHUB_TOKEN }}'
           script: |
+            const issueNumber = ${{ github.event.inputs.issue_number || inputs.issue_number }};
             const { data: issue } = await github.rest.issues.get({
               owner: context.repo.owner,
               repo: context.repo.repo,
-              issue_number: ${{ github.event.inputs.issue_number }},
+              issue_number: issueNumber,
             });
             core.setOutput('title', issue.title);
             core.setOutput('body', issue.body);
@@ -71,7 +78,7 @@ jobs:
         if: |-
           github.event_name == 'workflow_dispatch'
         env:
-          ISSUE_NUMBER_INPUT: '${{ github.event.inputs.issue_number }}'
+          ISSUE_NUMBER_INPUT: '${{ github.event.inputs.issue_number || inputs.issue_number }}'
           LABELS: '${{ steps.get_issue_data.outputs.labels }}'
         run: |
           if echo "${LABELS}" | grep -q 'area/'; then
@@ -86,6 +93,10 @@ jobs:
 
       - name: 'Generate GitHub App Token'
         id: 'generate_token'
+        env:
+          APP_ID: '${{ secrets.APP_ID }}'
+        if: |-
+          ${{ env.APP_ID != '' }}
         uses: 'actions/create-github-app-token@a8d616148505b5069dccd32f177bb87d7f39123b' # ratchet:actions/create-github-app-token@v2
         with:
           app-id: '${{ secrets.APP_ID }}'
@@ -96,7 +107,7 @@ jobs:
         id: 'get_labels'
         uses: 'actions/github-script@60a0d83039c74a4aee543508d2ffcb1c3799cdea'
         with:
-          github-token: '${{ steps.generate_token.outputs.token }}'
+          github-token: '${{ steps.generate_token.outputs.token || secrets.GITHUB_TOKEN }}'
           script: |-
             const { data: labels } = await github.rest.issues.listLabelsForRepo({
               owner: context.repo.owner,
@@ -127,7 +138,7 @@ jobs:
           ISSUE_BODY: >-
             ${{ github.event_name == 'workflow_dispatch' && steps.get_issue_data.outputs.body || github.event.issue.body }}
           ISSUE_NUMBER: >-
-            ${{ github.event_name == 'workflow_dispatch' && github.event.inputs.issue_number || github.event.issue.number }}
+            ${{ github.event_name == 'workflow_dispatch' && (github.event.inputs.issue_number || inputs.issue_number) || github.event.issue.number }}
           REPOSITORY: '${{ github.repository }}'
           AVAILABLE_LABELS: '${{ steps.get_labels.outputs.available_labels }}'
         with:
@@ -253,7 +264,7 @@ jobs:
           LABELS_OUTPUT: '${{ steps.gemini_issue_analysis.outputs.summary }}'
         uses: 'actions/github-script@60a0d83039c74a4aee543508d2ffcb1c3799cdea'
         with:
-          github-token: '${{ steps.generate_token.outputs.token }}'
+          github-token: '${{ steps.generate_token.outputs.token || secrets.GITHUB_TOKEN }}'
           script: |
             const rawOutput = process.env.LABELS_OUTPUT;
             core.info(`Raw output from model: ${rawOutput}`);
@@ -295,22 +306,6 @@ jobs:
             });
             core.info(`Successfully added labels for #${issueNumber}: ${labelsToAdd.join(', ')}`);
 
-            // Remove the 'status/need-triage' label
-            try {
-                await github.rest.issues.removeLabel({
-                    owner: context.repo.owner,
-                    repo: context.repo.repo,
-                    issue_number: issueNumber,
-                    name: 'status/need-triage'
-                });
-                core.info(`Successfully removed 'status/need-triage' label.`);
-            } catch (error) {
-                // If the label doesn't exist, the API call will throw a 404. We can ignore this.
-                if (error.status !== 404) {
-                    core.warning(`Failed to remove 'status/need-triage': ${error.message}`);
-                }
-            }
-
       - name: 'Post Issue Analysis Failure Comment'
         if: |-
           ${{ failure() && steps.gemini_issue_analysis.outcome == 'failure' }}
@@ -319,7 +314,7 @@ jobs:
           RUN_URL: '${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}'
         uses: 'actions/github-script@60a0d83039c74a4aee543508d2ffcb1c3799cdea'
         with:
-          github-token: '${{ steps.generate_token.outputs.token }}'
+          github-token: '${{ steps.generate_token.outputs.token || secrets.GITHUB_TOKEN }}'
           script: |-
             github.rest.issues.createComment({
               owner: context.repo.owner,

.github/workflows/gemini-automated-pr-size-labeler.yml

@@ -1,127 +0,0 @@
-name: 'Gemini Automated PR Labeler'
-
-on:
-  pull_request_target:
-    types: ['opened', 'reopened', 'synchronize']
-
-jobs:
-  label-pr:
-    timeout-minutes: 10
-    if: |-
-      ${{ github.repository == 'google-gemini/gemini-cli' }}
-    permissions:
-      pull-requests: 'write'
-      contents: 'read'
-      id-token: 'write'
-
-    concurrency:
-      group: '${{ github.workflow }}-${{ github.event.pull_request.number }}'
-      cancel-in-progress: true
-
-    runs-on: 'ubuntu-latest'
-
-    steps:
-      - name: 'Generate GitHub App Token'
-        id: 'generate_token'
-        uses: 'actions/create-github-app-token@df432ceedc7162793a195dd1713ff69aefc7379e' # v2
-        with:
-          app-id: '${{ secrets.APP_ID }}'
-          private-key: '${{ secrets.PRIVATE_KEY }}'
-          permission-pull-requests: 'write'
-
-      - name: 'Run Gemini PR size and complexity labeller'
-        uses: 'google-github-actions/run-gemini-cli@a3bf79042542528e91937b3a3a6fbc4967ee3c31' # Use the specific commit SHA
-        env:
-          GH_TOKEN: '${{ steps.generate_token.outputs.token }}'
-          PR_NUMBER: '${{ github.event.pull_request.number }}'
-          REPOSITORY: '${{ github.repository }}'
-        with:
-          gcp_workload_identity_provider: '${{ vars.GCP_WIF_PROVIDER }}'
-          gcp_project_id: '${{ vars.GOOGLE_CLOUD_PROJECT }}'
-          gcp_location: '${{ vars.GOOGLE_CLOUD_LOCATION }}'
-          gcp_service_account: '${{ vars.SERVICE_ACCOUNT_EMAIL }}'
-          gemini_api_key: '${{ secrets.GEMINI_API_KEY }}'
-          use_vertex_ai: '${{ vars.GOOGLE_GENAI_USE_VERTEXAI }}'
-          use_gemini_code_assist: '${{ vars.GOOGLE_GENAI_USE_GCA }}'
-          settings: |
-            {
-              "coreTools": [
-                "run_shell_command(gh pr diff)",
-                "run_shell_command(gh pr edit)",
-                "run_shell_command(gh pr comment)",
-                "run_shell_command(gh pr view)"
-              ],
-              "telemetry": {
-                "enabled": true,
-                "target": "gcp"
-              },
-              "sandbox": false
-            }
-          prompt: |
-            You are a Pull Request labeller and Feedback Assistant. Your primary goal is to improve review velocity and help maintainers prioritize their work by automatically labeling pull requests based on size and complexity, and providing guidance for overly large PRs.
-
-            Steps:
-            1. Retrieve Pull Request Information:
-               - Use `gh pr diff ${{ env.PR_NUMBER }} --repo ${{ env.REPOSITORY }}` to get the diff content.
-               - Parse the output from `gh pr diff` to determine the total lines of code added and deleted. Calculate `TOTAL_LINES_CHANGED`.
-
-            2. Determine Pull Request Size:
-               - Use `gh pr view ${{ env.PR_NUMBER }} --repo ${{ env.REPOSITORY }} --json labels` to get the current labels on the PR.
-               - Check the current labels and identify if any `size/*` labels already exist (e.g., `size/xs`, `size/s`, etc.).
-               - If an old `size/*` label is found and it is different from the newly calculated size, remove it using:
-                 `gh pr edit ${{ env.PR_NUMBER }} --repo ${{ env.REPOSITORY }} --remove-label "size-label-to-remove"`
-               - Based on `TOTAL_LINES_CHANGED`, select the appropriate new size label:
-                 - `size/xs`: < 10 lines changed
-                 - `size/s`: 10-50 lines changed
-                 - `size/m`: 51-200 lines changed
-                 - `size/l`: 201-1000 lines changed
-                 - `size/xl`: > 1000 lines changed
-               - Do not invent new size labels.
-               - Apply the newly determined `size/*` label to the pull request using:
-                 `gh pr edit ${{ env.PR_NUMBER }} --repo ${{ env.REPOSITORY }} --add-label "your-new-size-label"`
-
-            3. Analyze Pull Request Complexity:
-                - Perform Code Change Analysis: Examine the content of the code changes obtained from `gh pr diff ${{ env.PR_NUMBER }} --repo ${{ env.REPOSITORY }}`. Look for indicators of complexity such as:
-                 - Number of files changed (can be inferred from the diff headers).
-                 - Diversity of file types (e.g., changes across different languages, configuration files, documentation).
-                 - Presence of new external dependencies.
-                 - Introduction of new architectural components or significant refactoring.
-                 - Complexity of individual code changes (e.g., deeply nested logic, complex algorithms, extensive conditional statements).
-               - Apply Heuristic-based Complexity Assessment:
-                 - If the PR touches a small number of files with minor changes (e.g., typos, simple bug fixes, small feature additions), categorize it as `review/quick`.
-                 - If the PR involves changes across multiple files, introduces new features, significantly refactors existing code, or has a high line count (even within `size/l`), categorize it as `review/involved`.
-                 - Pay close attention to changes in critical or core modules as these inherently increase complexity.
-               - **Only use the labels `review/quick` or `review/involved` for complexity. Do not invent new complexity labels.**
-               - **Remove any previous `review/*` labels if they no longer apply, similar to the size label process.**
-               - Apply the determined `review/*` label to the pull request using:
-                 `gh pr edit ${{ env.PR_NUMBER }} --repo ${{ env.REPOSITORY }} --add-label "your-complexity-label"`
-
-            4. Handle Overly Large Pull Requests (`size/xl`):
-               - **Conditional Check:** If the pull request has been labeled `size/xl` (i.e., > 1000 lines of code changed) in Step 2, proceed to the next sub-step.
-               - **Post Constructive Comment:** Post a polite and helpful comment on the pull request using:
-                 `gh pr comment ${{ env.PR_NUMBER }} --repo ${{ env.REPOSITORY }} --body "Your comment here"`
-                 The comment body should be:
-                 """
-                 Thanks for your hard work on this pull request!
-
-                 This pull request is quite large, which can make it challenging and time-consuming for reviewers to go through thoroughly.
-
-                 To help us review it more efficiently and get your changes merged faster, we kindly request you consider breaking this into smaller, more focused pull requests. Each smaller PR should ideally address a single logical change or a small set of related changes.
-
-                 For example, you could separate out refactoring, new feature additions, and bug fixes into individual PRs. This makes it easier to understand, review, and test each component independently.
-
-                 We appreciate your understanding and cooperation. Feel free to reach out if you need any assistance with this!
-                 """
-
-            Guidelines:
-            - Automation Focus: All actions should be automated and not require manual intervention.
-            - Non-intrusive: The system should add labels and comments but not modify the code or close the pull request.
-            - Polite and Constructive: All communication, especially for large PRs, must be polite, encouraging, and constructive.
-            - Prioritize Clarity: The labels applied should clearly convey the PR's size and complexity to reviewers.
-            - Adhere to Defined Labels: Only use the specified `size/*` and `review/*` labels. Do not create or apply any other labels.
-            - Utilize `gh CLI`: Interact with GitHub using the `gh` command-line tool for diffing, label management, and commenting.
-            - Execute commands strictly as described in the steps. Do not invent new commands.
-            - In no case should you change other pull request that are not the one you are working on. Which can be found by using env.PR_NUMBER
-            - Execute each step that is defined in the steps section.
-            - In no case should you execute code from the pull request because this could be malicious code.
-            - If you fail to do this step log the errors you received

.github/workflows/gemini-scheduled-issue-triage.yml

@@ -1,12 +1,16 @@
 name: '📋 Gemini Scheduled Issue Triage'
 
 on:
+  issues:
+    types:
+      - 'opened'
+      - 'reopened'
   schedule:
     - cron: '0 * * * *' # Runs every hour
   workflow_dispatch:
 
 concurrency:
-  group: '${{ github.workflow }}'
+  group: '${{ github.workflow }}-${{ github.event.number || github.run_id }}'
   cancel-in-progress: true
 
 defaults:
@@ -35,7 +39,21 @@ jobs:
           private-key: '${{ secrets.PRIVATE_KEY }}'
           permission-issues: 'write'
 
+      - name: 'Get issue from event'
+        if: |-
+          ${{ github.event_name == 'issues' }}
+        id: 'get_issue_from_event'
+        env:
+          ISSUE_EVENT: '${{ toJSON(github.event.issue) }}'
+        run: |
+          set -euo pipefail
+          ISSUE_JSON=$(echo "$ISSUE_EVENT" | jq -c '[{number: .number, title: .title, body: .body}]')
+          echo "issues_to_triage=${ISSUE_JSON}" >> "${GITHUB_OUTPUT}"
+          echo "✅ Found issue #${{ github.event.issue.number }} from event to triage! 🎯"
+
       - name: 'Find untriaged issues'
+        if: |-
+          ${{ github.event_name == 'schedule' || github.event_name == 'workflow_dispatch' }}
         id: 'find_issues'
         env:
           GITHUB_TOKEN: '${{ steps.generate_token.outputs.token }}'
@@ -43,22 +61,26 @@ jobs:
         run: |-
           set -euo pipefail
 
-          echo '🔍 Finding issues without labels...'
-          NO_LABEL_ISSUES="$(gh issue list --repo "${GITHUB_REPOSITORY}" \
-            --search 'is:open is:issue no:label' --limit 10 --json number,title,body)"
+          echo '🔍 Finding issues missing area labels...'
+          NO_AREA_ISSUES="$(gh issue list --repo "${GITHUB_REPOSITORY}" \
+            --search 'is:open is:issue -label:area/core -label:area/agent -label:area/enterprise -label:area/non-interactive -label:area/security -label:area/platform -label:area/extensions -label:area/unknown' --limit 100 --json number,title,body)"
 
-          echo '🏷️ Finding issues that need triage...'
-          NEED_TRIAGE_ISSUES="$(gh issue list --repo "${GITHUB_REPOSITORY}" \
-            --search "is:open is:issue label:\"status/need-triage\"" --limit 10 --json number,title,body)"
+          echo '🔍 Finding issues missing kind labels...'
+          NO_KIND_ISSUES="$(gh issue list --repo "${GITHUB_REPOSITORY}" \
+            --search 'is:open is:issue -label:kind/bug -label:kind/enhancement -label:kind/customer-issue -label:kind/question' --limit 100 --json number,title,body)"
+
+          echo '🏷️ Finding issues missing priority labels...'
+          NO_PRIORITY_ISSUES="$(gh issue list --repo "${GITHUB_REPOSITORY}" \
+            --search 'is:open is:issue -label:priority/p0 -label:priority/p1 -label:priority/p2 -label:priority/p3 -label:priority/unknown' --limit 100 --json number,title,body)"
 
           echo '🔄 Merging and deduplicating issues...'
-          ISSUES="$(echo "${NO_LABEL_ISSUES}" "${NEED_TRIAGE_ISSUES}" | jq -c -s 'add | unique_by(.number)')"
+          ISSUES="$(echo "${NO_AREA_ISSUES}" "${NO_KIND_ISSUES}" "${NO_PRIORITY_ISSUES}" | jq -c -s 'add | unique_by(.number)')"
 
           echo '📝 Setting output for GitHub Actions...'
           echo "issues_to_triage=${ISSUES}" >> "${GITHUB_OUTPUT}"
 
           ISSUE_COUNT="$(echo "${ISSUES}" | jq 'length')"
-          echo "✅ Found ${ISSUE_COUNT} issues to triage! 🎯"
+          echo "✅ Found ${ISSUE_COUNT} unique issues to triage! 🎯"
 
       - name: 'Get Repository Labels'
         id: 'get_labels'
@@ -77,12 +99,13 @@ jobs:
 
       - name: 'Run Gemini Issue Analysis'
         if: |-
-          ${{ steps.find_issues.outputs.issues_to_triage != '[]' }}
+          (steps.get_issue_from_event.outputs.issues_to_triage != '' && steps.get_issue_from_event.outputs.issues_to_triage != '[]') ||
+          (steps.find_issues.outputs.issues_to_triage != '' && steps.find_issues.outputs.issues_to_triage != '[]')
         uses: 'google-github-actions/run-gemini-cli@a3bf79042542528e91937b3a3a6fbc4967ee3c31' # ratchet:google-github-actions/run-gemini-cli@v0
         id: 'gemini_issue_analysis'
         env:
           GITHUB_TOKEN: '' # Do not pass any auth token here since this runs on untrusted inputs
-          ISSUES_TO_TRIAGE: '${{ steps.find_issues.outputs.issues_to_triage }}'
+          ISSUES_TO_TRIAGE: '${{ steps.get_issue_from_event.outputs.issues_to_triage || steps.find_issues.outputs.issues_to_triage }}'
           REPOSITORY: '${{ github.repository }}'
           AVAILABLE_LABELS: '${{ steps.get_labels.outputs.available_labels }}'
         with:
@@ -116,38 +139,32 @@ jobs:
             1. You are only able to use the echo command. Review the available labels in the environment variable: "${AVAILABLE_LABELS}".
             2. Check environment variable for issues to triage: $ISSUES_TO_TRIAGE (JSON array of issues)
             3. Review the issue title, body and any comments provided in the environment variables.
-            4. Identify the most relevant labels from the existing labels, focusing on kind/*, area/*, sub-area/* and priority/*.
-            5. If the issue already has area/ label, dont try to change it. Similarly, if the issue already has a kind/ label don't change it. And if the issue already has a priority/ label do not change it for example:
-               If an issue has area/core and kind/bug  you will only add a priority/ label.
-               Instead if an issue has no labels, you will could add one lable of each kind.
+            4. Identify the most relevant labels from the existing labels, specifically focusing on area/*, kind/* and priority/*.
+            5. Label Policy:
+               - If the issue already has a kind/ label, do not change it.
+               - If the issue already has a priority/ label, do not change it.
+               - If the issue already has an area/ label, do not change it.
+               - If any of these are missing, select exactly ONE appropriate label for the missing category.
             6. Identify other applicable labels based on the issue content, such as status/*, help wanted, good first issue, etc.
-            7. For area/* and kind/* limit yourself to only the single most applicable label in each case.
-            8. Give me a single short explanation about why you are selecting each label in the process.
-            9. Output a JSON array of objects, each containing the issue number
+            7. Give me a single short explanation about why you are selecting each label in the process.
+            8. Output a JSON array of objects, each containing the issue number
                and the labels to add and remove, along with an explanation. For example:
                ```
                [
                  {
                    "issue_number": 123,
-                   "labels_to_add": ["kind/bug", "priority/p2"],
+                   "labels_to_add": ["area/core", "kind/bug", "priority/p2"],
                    "labels_to_remove": ["status/need-triage"],
-                   "explanation": "This issue is a bug that needs to be addressed with medium priority."
-                 },
-                 {
-                   "issue_number": 456,
-                   "labels_to_add": ["kind/enhancement"],
-                   "labels_to_remove": [],
-                   "explanation": "This issue is an enhancement request that could improve the user experience."
+                   "explanation": "This issue is a UI bug that needs to be addressed with medium priority."
                  }
                ]
                ```
               If an issue cannot be classified, do not include it in the output array.
-            10. For each issue please check if CLI version is present, this is usually in the output of the /about command and will look like 0.1.5
+            9. For each issue please check if CLI version is present, this is usually in the output of the /about command and will look like 0.1.5
               - Anything more than 6 versions older than the most recent should add the status/need-retesting label
-            11. If you see that the issue doesn't look like it has sufficient information recommend the status/need-information label and leave a comment politely requesting the relevant information, eg.. if repro steps are missing request for repro steps. if version information is missing request for version information into the explanation section below.
-              - After identifying appropriate labels to an issue, add "status/need-triage" label to labels_to_remove in the output.
-            12. If you think an issue might be a Priority/P0 do not apply the priority/p0 label. Instead apply a status/manual-triage label and include a note in your explanation.
-            13. If you are uncertain and have not been able to apply one each of kind/, area/ and priority/ , apply the status/manual-triage label.
+            10. If you see that the issue doesn't look like it has sufficient information recommend the status/need-information label and leave a comment politely requesting the relevant information, eg.. if repro steps are missing request for repro steps. if version information is missing request for version information into the explanation section below.
+            11. If you think an issue might be a Priority/P0 do not apply the priority/p0 label. Instead apply a status/manual-triage label and include a note in your explanation.
+            12. If you are uncertain about a category, use the area/unknown, kind/question, or priority/unknown labels as appropriate. If you are extremely uncertain, apply the status/manual-triage label.
 
             ## Guidelines
 
@@ -157,100 +174,46 @@ jobs:
               - Do not add comments or modify the issue content.
               - Do not remove the following labels maintainer, help wanted or good first issue.
               - Triage only the current issue.
-              - Identify only one area/ label
+              - Identify only one area/ label.
               - Identify only one kind/ label (Do not apply kind/duplicate or kind/parent-issue)
-              - Identify all applicable sub-area/* and priority/* labels based on the issue content. It's ok to have multiple of these.
+              - Identify only one priority/ label.
               - Once you categorize the issue if it needs information bump down the priority by 1 eg.. a p0 would become a p1 a p1 would become a p2. P2 and P3 can stay as is in this scenario.
-            Categorization Guidelines:
-            P0: Critical / Blocker
-              - A P0 bug is a catastrophic failure that demands immediate attention.
-              - To be a P0 it means almost all users are running into this issue and it is blocking users from being able to use the product.
-              - You would see this in the form of many comments from different developers on the bug.
-              - It represents a complete showstopper for a significant portion of users or for the development process itself.
-            Impact:
-              - Blocks development or testing for the entire team.
-              - Major security vulnerability that could compromise user data or system integrity.
-              - Causes data loss or corruption with no workaround.
-              - Crashes the application or makes a core feature completely unusable for all or most users in a production environment. Will it cause severe quality degration?
-              - Is it preventing contributors from contributing to the repository or is it a release blocker?
-              Qualifier: Is the main function of the software broken?
-              Example: The gemini auth login command fails with an unrecoverable error, preventing any user from authenticating and using the rest of the CLI.
-            P1: High
-              - A P1 bug is a serious issue that significantly degrades the user experience or impacts a core feature.
-              - While not a complete blocker, it's a major problem that needs a fast resolution. Feature requests are almost never P1.
-              - Once again this would be affecting many users.
-              - You would see this in the form of comments from different developers on the bug.
-              Impact:
-              - A core feature is broken or behaving incorrectly for a large number of users or large number of use cases.
-              - Review the bug details and comments to try figure out if this issue affects a large set of use cases or if it's a narrow set of use cases.
-              - Severe performance degradation making the application frustratingly slow.
-              - No straightforward workaround exists, or the workaround is difficult and non-obvious.
-              Qualifier: Is a key feature unusable or giving very wrong results?
-              Example: Gemini CLI enters a loop when making read-many-files tool call. I am unable to break out of the loop and gemini doesn't follow instructions subsequently.
-            P2: Medium
-              - A P2 bug is a moderately impactful issue. It's a noticeable problem but doesn't prevent the use of the software's main functionality.
-              Impact:
-              - Affects a non-critical feature or a smaller, specific subset of users.
-              - An inconvenient but functional workaround is available and easy to execute.
-              - Noticeable UI/UX problems that don't break functionality but look unprofessional (e.g., elements are misaligned or overlapping).
-              Qualifier: Is it an annoying but non-blocking problem?
-              Example: An error message is unclear or contains a typo, causing user confusion but not halting their workflow.
-            P3: Low
-              - A P3 bug is a minor, low-impact issue that is trivial or cosmetic. It has little to no effect on the overall functionality of the application.
-              Impact:
-              - Minor cosmetic issues like color inconsistencies, typos in documentation, or slight alignment problems on a non-critical page.
-              - An edge-case bug that is very difficult to reproduce and affects a tiny fraction of users.
-              Qualifier: Is it a "nice-to-fix" issue?
-              Example: Spelling mistakes etc.
+
+            Categorization Guidelines (Priority):
+            P0 - Urgent Blocking Issues:
+              - DO NOT APPLY THIS LABEL AUTOMATICALLY. Use status/manual-triage instead.
+              - Definition: Urgent, block a significant percentage of the user base, and prevent frequent use of the Gemini CLI.
+              - This includes core stability blockers (e.g., authentication failures, broken upgrades), critical crashes, and P0 security vulnerabilities.
+              - Impact: Blocks development or testing for the entire team; Major security vulnerability; Causes data loss or corruption with no workaround; Crashes the application or makes a core feature completely unusable for all or most users.
+              - Qualifier: Is the main function of the software broken?
+            P1 - High-Impact Issues:
+              - Definition: Affect a large number of users, blocking them from using parts of the Gemini CLI, or make the CLI frequently unusable even with workarounds available.
+              - Impact: A core feature is broken or behaving incorrectly for a large number of users or use cases; Severe performance degradation; No straightforward workaround exists.
+              - Qualifier: Is a key feature unusable or giving very wrong results?
+            P2 - Significant Issues:
+              - Definition: Affect some users significantly, such as preventing the use of certain features or authentication types.
+              - Can also be issues that many users complain about, causing annoyance or hindering daily use.
+              - Impact: Affects a non-critical feature or a smaller, specific subset of users; An inconvenient but functional workaround is available; Noticeable UI/UX problems that look unprofessional.
+              - Qualifier: Is it an annoying but non-blocking problem?
+            P3 - Low-Impact Issues:
+              - Definition: Typically usability issues that cause annoyance to a limited user base.
+              - Includes feature requests that could be addressed in the near future and may be suitable for community contributions.
+              - Impact: Minor cosmetic issues; An edge-case bug that is very difficult to reproduce and affects a tiny fraction of users.
+              - Qualifier: Is it a "nice-to-fix" issue?
+
+            Categorization Guidelines (Area):
+            area/agent: Core Agent, Tools, Memory, Sub-Agents, Hooks, Agent Quality
+            area/core: User Interface, OS Support, Core Functionality
+            area/enterprise: Telemetry, Policy, Quota / Licensing
+            area/extensions: Gemini CLI extensions capability
+            area/non-interactive: GitHub Actions, SDK, 3P Integrations, Shell Scripting, Command line automation
+            area/platform: Build infra, Release mgmt, Testing, Eval infra, Capacity, Quota mgmt
+            area/security: security related issues
+
             Additional Context:
-              - If users are talking about issues where the model gets downgraded from pro to flash then i want you to categorize that as a performance issue
-              - This product is designed to use different models eg.. using pro, downgrading to flash etc.
-              - When users report that they dont expect the model to change those would be categorized as feature requests.
-            Definition of Areas
-            area/ux:
-              - Issues concerning user-facing elements like command usability, interactive features, help docs, and perceived performance.
-              - I am seeing my screen flicker when using Gemini CLI
-              - I am seeing the output malformed
-              - Theme changes aren't taking effect
-              - My keyboard inputs arent' being recognzied
-            area/platform:
-              - Issues related to installation, packaging, OS compatibility (Windows, macOS, Linux), and the underlying CLI framework.
-            area/background: Issues related to long-running background tasks, daemons, and autonomous or proactive agent features.
-            area/models:
-              - i am not getting a response that is reasonable or expected. this can include things like
-              - I am calling a tool and the tool is not performing as expected.
-              - i am expecting a tool to be called and it is not getting called ,
-              - Including experience when using
-              - built-in tools (e.g., web search, code interpreter, read file, writefile, etc..),
-              - Function calling issues should be under this area
-              - i am getting responses from the model that are malformed.
-              - Issues concerning Gemini quality of response and inference,
-              - Issues talking about unnecessary token consumption.
-              - Issues talking about Model getting stuck in a loop be watchful as this could be the root cause for issues that otherwise seem like model performance issues.
-              - Memory compression
-              - unexpected responses,
-              - poor quality of generated code
-            area/tools:
-              - These are primarily issues related to Model Context Protocol
-              - These are issues that mention MCP support
-              - feature requests asking for support for new tools.
-            area/core:
-              - Issues with fundamental components like command parsing, configuration management, session state, and the main API client logic. Introducing multi-modality
-            area/contribution:
-              - Issues related to improving the developer contribution experience, such as CI/CD pipelines, build scripts, and test automation infrastructure.
-            area/authentication:
-              -  Issues related to user identity, login flows, API key handling, credential storage, and access token management, unable to sign in selecting wrong authentication path etc..
-            area/security-privacy:
-              - Issues concerning vulnerability patching, dependency security, data sanitization, privacy controls, and preventing unauthorized data access.
-            area/extensibility:
-              - Issues related to the plugin system, extension APIs, or making the CLI's functionality available in other applications, github actions, ide support etc..
-            area/performance:
-              - Issues focused on model performance
-              - Issues with running out of capacity,
-              - 429 errors etc..
-              - could also pertain to latency,
-              - other general software performance like, memory usage, CPU consumption, and algorithmic efficiency.
-              - Switching models from one to the other unexpectedly.
+            - If users are talking about issues where the model gets downgraded from pro to flash then i want you to categorize that as a performance issue.
+            - This product is designed to use different models eg.. using pro, downgrading to flash etc.
+            - When users report that they dont expect the model to change those would be categorized as feature requests.
 
       - name: 'Apply Labels to Issues'
         if: |-
@@ -300,24 +263,6 @@ jobs:
                 core.info(`Successfully added labels for #${issueNumber}: ${labelsToAdd.join(', ')}${explanation}`);
               }
 
-              if (entry.labels_to_remove && entry.labels_to_remove.length > 0) {
-                for (const label of entry.labels_to_remove) {
-                  try {
-                    await github.rest.issues.removeLabel({
-                      owner: context.repo.owner,
-                      repo: context.repo.repo,
-                      issue_number: issueNumber,
-                      name: label
-                    });
-                  } catch (error) {
-                    if (error.status !== 404) {
-                      throw error;
-                    }
-                  }
-                }
-                core.info(`Successfully removed labels for #${issueNumber}: ${entry.labels_to_remove.join(', ')}`);
-              }
-
               if (entry.explanation) {
                 await github.rest.issues.createComment({
                   owner: context.repo.owner,

.github/workflows/gemini-scheduled-pr-triage.yml

@@ -39,3 +39,7 @@ jobs:
           GITHUB_REPOSITORY: '${{ github.repository }}'
         run: |-
           ./.github/scripts/pr-triage.sh
+          # If prs_needing_comment is empty, set it to [] explicitly for downstream steps
+          if [[ -z "$(grep 'prs_needing_comment' "${GITHUB_OUTPUT}" | cut -d'=' -f2-)" ]]; then
+            echo "prs_needing_comment=[]" >> "${GITHUB_OUTPUT}"
+          fi

.github/workflows/gemini-scheduled-stale-issue-closer.yml

@@ -79,8 +79,14 @@ jobs:
                 continue;
               }
 
-              // Skip if it has a maintainer label
-              if (issue.labels.some(label => label.name.toLowerCase().includes('maintainer'))) {
+              // Skip if it has a maintainer, help wanted, or Public Roadmap label
+              const rawLabels = issue.labels.map((l) => l.name);
+              const lowercaseLabels = rawLabels.map((l) => l.toLowerCase());
+              if (
+                lowercaseLabels.some((l) => l.includes('maintainer')) ||
+                lowercaseLabels.includes('help wanted') ||
+                rawLabels.includes('🗓️ Public Roadmap')
+              ) {
                 continue;
               }
 

.github/workflows/gemini-scheduled-stale-pr-closer.yml

@@ -0,0 +1,205 @@
+name: 'Gemini Scheduled Stale PR Closer'
+
+on:
+  schedule:
+    - cron: '0 2 * * *' # Every day at 2 AM UTC
+  pull_request:
+    types: ['opened', 'edited']
+  workflow_dispatch:
+    inputs:
+      dry_run:
+        description: 'Run in dry-run mode'
+        required: false
+        default: false
+        type: 'boolean'
+
+jobs:
+  close-stale-prs:
+    if: "github.repository == 'google-gemini/gemini-cli'"
+    runs-on: 'ubuntu-latest'
+    permissions:
+      pull-requests: 'write'
+      issues: 'write'
+    steps:
+      - name: 'Generate GitHub App Token'
+        id: 'generate_token'
+        uses: 'actions/create-github-app-token@v1'
+        with:
+          app-id: '${{ secrets.APP_ID }}'
+          private-key: '${{ secrets.PRIVATE_KEY }}'
+          owner: '${{ github.repository_owner }}'
+          repositories: 'gemini-cli'
+
+      - name: 'Process Stale PRs'
+        uses: 'actions/github-script@v7'
+        env:
+          DRY_RUN: '${{ inputs.dry_run }}'
+        with:
+          github-token: '${{ steps.generate_token.outputs.token }}'
+          script: |
+            const dryRun = process.env.DRY_RUN === 'true';
+            const thirtyDaysAgo = new Date();
+            thirtyDaysAgo.setDate(thirtyDaysAgo.getDate() - 30);
+
+            // 1. Fetch maintainers for verification
+            let maintainerLogins = new Set();
+            try {
+              const members = await github.paginate(github.rest.teams.listMembersInOrg, {
+                org: context.repo.owner,
+                team_slug: 'gemini-cli-maintainers'
+              });
+              maintainerLogins = new Set(members.map(m => m.login));
+            } catch (e) {
+              core.warning('Failed to fetch team members');
+            }
+
+            const isMaintainer = (login, assoc) => {
+              if (maintainerLogins.size > 0) return maintainerLogins.has(login);
+              return ['OWNER', 'MEMBER', 'COLLABORATOR'].includes(assoc);
+            };
+
+            // 2. Determine which PRs to check
+            let prs = [];
+            if (context.eventName === 'pull_request') {
+              const { data: pr } = await github.rest.pulls.get({
+                owner: context.repo.owner,
+                repo: context.repo.repo,
+                pull_number: context.payload.pull_request.number
+              });
+              prs = [pr];
+            } else {
+              prs = await github.paginate(github.rest.pulls.list, {
+                owner: context.repo.owner,
+                repo: context.repo.repo,
+                state: 'open',
+                per_page: 100
+              });
+            }
+
+            for (const pr of prs) {
+              const maintainerPr = isMaintainer(pr.user.login, pr.author_association);
+              const isBot = pr.user.type === 'Bot' || pr.user.login.endsWith('[bot]');
+
+              // Detection Logic for Linked Issues
+              // Check 1: Official GitHub "Closing Issue" link (GraphQL)
+              const linkedIssueQuery = `query($owner:String!, $repo:String!, $number:Int!) {
+                repository(owner:$owner, name:$repo) {
+                  pullRequest(number:$number) {
+                    closingIssuesReferences(first: 1) { totalCount }
+                  }
+                }
+              }`;
+
+              let hasClosingLink = false;
+              try {
+                const res = await github.graphql(linkedIssueQuery, {
+                  owner: context.repo.owner, repo: context.repo.repo, number: pr.number
+                });
+                hasClosingLink = res.repository.pullRequest.closingIssuesReferences.totalCount > 0;
+              } catch (e) {}
+
+              // Check 2: Regex for mentions (e.g., "Related to #123", "Part of #123", "#123")
+              // We check for # followed by numbers or direct URLs to issues.
+              const body = pr.body || '';
+              const mentionRegex = /(?:#|https:\/\/github\.com\/[^\/]+\/[^\/]+\/issues\/)(\d+)/i;
+              const hasMentionLink = mentionRegex.test(body);
+
+              const hasLinkedIssue = hasClosingLink || hasMentionLink;
+
+              // Logic for Closed PRs (Auto-Reopen)
+              if (pr.state === 'closed' && context.eventName === 'pull_request' && context.payload.action === 'edited') {
+                if (hasLinkedIssue) {
+                  core.info(`PR #${pr.number} now has a linked issue. Reopening.`);
+                  if (!dryRun) {
+                    await github.rest.pulls.update({
+                      owner: context.repo.owner,
+                      repo: context.repo.repo,
+                      pull_number: pr.number,
+                      state: 'open'
+                    });
+                    await github.rest.issues.createComment({
+                      owner: context.repo.owner,
+                      repo: context.repo.repo,
+                      issue_number: pr.number,
+                      body: "Thank you for linking an issue! This pull request has been automatically reopened."
+                    });
+                  }
+                }
+                continue;
+              }
+
+              // Logic for Open PRs (Immediate Closure)
+              if (pr.state === 'open' && !maintainerPr && !hasLinkedIssue && !isBot) {
+                core.info(`PR #${pr.number} is missing a linked issue. Closing.`);
+                if (!dryRun) {
+                  await github.rest.issues.createComment({
+                    owner: context.repo.owner,
+                    repo: context.repo.repo,
+                    issue_number: pr.number,
+                    body: "Hi there! Thank you for your contribution to Gemini CLI. \n\nTo improve our contribution process and better track changes, we now require all pull requests to be associated with an existing issue, as announced in our [recent discussion](https://github.com/google-gemini/gemini-cli/discussions/16706) and as detailed in our [CONTRIBUTING.md](https://github.com/google-gemini/gemini-cli/blob/main/CONTRIBUTING.md#1-link-to-an-existing-issue).\n\nThis pull request is being closed because it is not currently linked to an issue. **Once you have updated the description of this PR to link an issue (e.g., by adding `Fixes #123` or `Related to #123`), it will be automatically reopened.**\n\n**How to link an issue:**\nAdd a keyword followed by the issue number (e.g., `Fixes #123`) in the description of your pull request. For more details on supported keywords and how linking works, please refer to the [GitHub Documentation on linking pull requests to issues](https://docs.github.com/en/issues/tracking-your-work-with-issues/linking-a-pull-request-to-an-issue).\n\nThank you for your understanding and for being a part of our community!"
+                  });
+                  await github.rest.pulls.update({
+                    owner: context.repo.owner,
+                    repo: context.repo.repo,
+                    pull_number: pr.number,
+                    state: 'closed'
+                  });
+                }
+                continue;
+              }
+
+              // Staleness check (Scheduled runs only)
+              if (pr.state === 'open' && context.eventName !== 'pull_request') {
+                const labels = pr.labels.map(l => l.name.toLowerCase());
+                if (labels.includes('help wanted') || labels.includes('🔒 maintainer only')) continue;
+
+                let lastActivity = new Date(0);
+                try {
+                  const reviews = await github.paginate(github.rest.pulls.listReviews, {
+                    owner: context.repo.owner,
+                    repo: context.repo.repo,
+                    pull_number: pr.number
+                  });
+                  for (const r of reviews) {
+                    if (isMaintainer(r.user.login, r.author_association)) {
+                      const d = new Date(r.submitted_at || r.updated_at);
+                      if (d > lastActivity) lastActivity = d;
+                    }
+                  }
+                  const comments = await github.paginate(github.rest.issues.listComments, {
+                    owner: context.repo.owner,
+                    repo: context.repo.repo,
+                    issue_number: pr.number
+                  });
+                  for (const c of comments) {
+                    if (isMaintainer(c.user.login, c.author_association)) {
+                      const d = new Date(c.updated_at);
+                      if (d > lastActivity) lastActivity = d;
+                    }
+                  }
+                } catch (e) {}
+
+                if (maintainerPr) {
+                  const d = new Date(pr.created_at);
+                  if (d > lastActivity) lastActivity = d;
+                }
+
+                if (lastActivity < thirtyDaysAgo) {
+                  core.info(`PR #${pr.number} is stale.`);
+                  if (!dryRun) {
+                    await github.rest.issues.createComment({
+                      owner: context.repo.owner,
+                      repo: context.repo.repo,
+                      issue_number: pr.number,
+                      body: "Hi there! Thank you for your contribution to Gemini CLI. We really appreciate the time and effort you've put into this pull request.\n\nTo keep our backlog manageable and ensure we're focusing on current priorities, we are closing pull requests that haven't seen maintainer activity for 30 days. Currently, the team is prioritizing work associated with **🔒 maintainer only** or **help wanted** issues.\n\nIf you believe this change is still critical, please feel free to comment with updated details. Otherwise, we encourage contributors to focus on open issues labeled as **help wanted**. Thank you for your understanding!"
+                    });
+                    await github.rest.pulls.update({
+                      owner: context.repo.owner,
+                      repo: context.repo.repo,
+                      pull_number: pr.number,
+                      state: 'closed'
+                    });
+                  }
+                }
+              }
+            }

.github/workflows/issue-opened-labeler.yml

@@ -0,0 +1,46 @@
+name: '🏷️ Issue Opened Labeler'
+
+on:
+  issues:
+    types:
+      - 'opened'
+
+jobs:
+  label-issue:
+    runs-on: 'ubuntu-latest'
+    if: |-
+      ${{ github.repository == 'google-gemini/gemini-cli' || github.repository == 'google-gemini/maintainers-gemini-cli' }}
+    steps:
+      - name: 'Generate GitHub App Token'
+        id: 'generate_token'
+        env:
+          APP_ID: '${{ secrets.APP_ID }}'
+        if: |-
+          ${{ env.APP_ID != '' }}
+        uses: 'actions/create-github-app-token@a8d616148505b5069dccd32f177bb87d7f39123b' # ratchet:actions/create-github-app-token@v2
+        with:
+          app-id: '${{ secrets.APP_ID }}'
+          private-key: '${{ secrets.PRIVATE_KEY }}'
+
+      - name: 'Add need-triage label'
+        uses: 'actions/github-script@60a0d83039c74a4aee543508d2ffcb1c3799cdea'
+        with:
+          github-token: '${{ steps.generate_token.outputs.token || secrets.GITHUB_TOKEN }}'
+          script: |-
+            const { data: issue } = await github.rest.issues.get({
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              issue_number: context.issue.number,
+            });
+
+            const hasLabel = issue.labels.some(l => l.name === 'status/need-triage');
+            if (!hasLabel) {
+              await github.rest.issues.addLabels({
+                owner: context.repo.owner,
+                repo: context.repo.repo,
+                issue_number: context.issue.number,
+                labels: ['status/need-triage']
+              });
+            } else {
+              core.info('Issue already has status/need-triage label. Skipping.');
+            }

.github/workflows/label-backlog-child-issues.yml

@@ -3,38 +3,55 @@ name: 'Label Child Issues for Project Rollup'
 on:
   issues:
     types: ['opened', 'edited', 'reopened']
+  schedule:
+    - cron: '0 * * * *' # Run every hour
+  workflow_dispatch:
+
+permissions:
+  issues: 'write'
+  contents: 'read'
 
 jobs:
+  # Event-based: Quick reaction to new/edited issues in THIS repo
   labeler:
+    if: "github.event_name == 'issues'"
     runs-on: 'ubuntu-latest'
-    permissions:
-      issues: 'write'
     steps:
-      - name: 'Check for Parent Workstream and Apply Label'
-        uses: 'actions/github-script@v7'
+      - name: 'Checkout'
+        uses: 'actions/checkout@v4'
+
+      - name: 'Setup Node.js'
+        uses: 'actions/setup-node@v4'
         with:
-          script: |
-            const issue = context.payload.issue;
-            const labelToAdd = 'workstream-rollup';
+          node-version: '20'
+          cache: 'npm'
 
-            // --- Define the FULL URLs of the allowed parent workstreams ---
-            const allowedParentUrls = [
-              'https://api.github.com/repos/google-gemini/gemini-cli/issues/15374',
-              'https://api.github.com/repos/google-gemini/gemini-cli/issues/15456',
-              'https://api.github.com/repos/google-gemini/gemini-cli/issues/15324'
-            ];
+      - name: 'Install Dependencies'
+        run: 'npm ci'
 
-            // Check if the issue has a parent_issue_url and if it's in our allowed list.
-            if (issue && issue.parent_issue_url && allowedParentUrls.includes(issue.parent_issue_url)) {
-              console.log(`SUCCESS: Issue #${issue.number} is a child of a target workstream (${issue.parent_issue_url}). Adding label.`);
-              await github.rest.issues.addLabels({
-                owner: context.repo.owner,
-                repo: context.repo.repo,
-                issue_number: issue.number,
-                labels: [labelToAdd]
-              });
-            } else if (issue && issue.parent_issue_url) {
-              console.log(`FAILURE: Issue #${issue.number} has a parent, but it's not a target workstream. Parent URL: ${issue.parent_issue_url}`);
-            } else {
-              console.log(`FAILURE: Issue #${issue.number} is not a child of any issue. No action taken.`);
-            }
+      - name: 'Run Multi-Repo Sync Script'
+        env:
+          GITHUB_TOKEN: '${{ secrets.GITHUB_TOKEN }}'
+        run: 'node .github/scripts/sync-maintainer-labels.cjs'
+
+  # Scheduled/Manual: Recursive sync across multiple repos
+  sync-maintainer-labels:
+    if: "github.event_name == 'schedule' || github.event_name == 'workflow_dispatch'"
+    runs-on: 'ubuntu-latest'
+    steps:
+      - name: 'Checkout'
+        uses: 'actions/checkout@v4'
+
+      - name: 'Setup Node.js'
+        uses: 'actions/setup-node@v4'
+        with:
+          node-version: '20'
+          cache: 'npm'
+
+      - name: 'Install Dependencies'
+        run: 'npm ci'
+
+      - name: 'Run Multi-Repo Sync Script'
+        env:
+          GITHUB_TOKEN: '${{ secrets.GITHUB_TOKEN }}'
+        run: 'node .github/scripts/sync-maintainer-labels.cjs'

.github/workflows/label-workstream-rollup.yml

@@ -0,0 +1,174 @@
+name: 'Label Workstream Rollup'
+
+on:
+  issues:
+    types: ['opened', 'edited', 'reopened']
+  schedule:
+    - cron: '0 * * * *'
+  workflow_dispatch:
+
+jobs:
+  labeler:
+    runs-on: 'ubuntu-latest'
+    permissions:
+      issues: 'write'
+    steps:
+      - name: 'Check for Parent Workstream and Apply Label'
+        uses: 'actions/github-script@v7'
+        with:
+          script: |
+            const labelToAdd = 'workstream-rollup';
+
+            // Allow-list of parent issue URLs
+            const allowedParentUrls = [
+              'https://github.com/google-gemini/gemini-cli/issues/15374',
+              'https://github.com/google-gemini/gemini-cli/issues/15456',
+              'https://github.com/google-gemini/gemini-cli/issues/15324',
+              'https://github.com/google-gemini/gemini-cli/issues/17202',
+              'https://github.com/google-gemini/gemini-cli/issues/17203'
+            ];
+
+            // Single issue processing (for event triggers)
+            async function processSingleIssue(owner, repo, number) {
+              const query = `
+                query($owner:String!, $repo:String!, $number:Int!) {
+                  repository(owner:$owner, name:$repo) {
+                    issue(number:$number) {
+                      number
+                      parent {
+                        url
+                        parent {
+                          url
+                          parent {
+                            url
+                            parent {
+                              url
+                              parent {
+                                url
+                              }
+                            }
+                          }
+                        }
+                      }
+                    }
+                  }
+                }
+              `;
+              try {
+                const result = await github.graphql(query, { owner, repo, number });
+
+                if (!result || !result.repository || !result.repository.issue) {
+                  console.log(`Issue #${number} not found or data missing.`);
+                  return;
+                }
+
+                const issue = result.repository.issue;
+                await checkAndLabel(issue, owner, repo);
+              } catch (error) {
+                console.error(`Failed to process issue #${number}:`, error);
+                throw error; // Re-throw to be caught by main execution
+              }
+            }
+
+            // Bulk processing (for schedule/dispatch)
+            async function processAllOpenIssues(owner, repo) {
+              const query = `
+                query($owner:String!, $repo:String!, $cursor:String) {
+                  repository(owner:$owner, name:$repo) {
+                    issues(first: 100, states: OPEN, after: $cursor) {
+                      pageInfo {
+                        hasNextPage
+                        endCursor
+                      }
+                      nodes {
+                        number
+                        parent {
+                          url
+                          parent {
+                            url
+                            parent {
+                              url
+                              parent {
+                                url
+                                parent {
+                                  url
+                                }
+                              }
+                            }
+                          }
+                        }
+                      }
+                    }
+                  }
+                }
+              `;
+
+              let hasNextPage = true;
+              let cursor = null;
+
+              while (hasNextPage) {
+                try {
+                  const result = await github.graphql(query, { owner, repo, cursor });
+
+                  if (!result || !result.repository || !result.repository.issues) {
+                     console.error('Invalid response structure from GitHub API');
+                     break;
+                  }
+
+                  const issues = result.repository.issues.nodes || [];
+
+                  console.log(`Processing batch of ${issues.length} issues...`);
+                  for (const issue of issues) {
+                    await checkAndLabel(issue, owner, repo);
+                  }
+
+                  hasNextPage = result.repository.issues.pageInfo.hasNextPage;
+                  cursor = result.repository.issues.pageInfo.endCursor;
+                } catch (error) {
+                  console.error('Failed to fetch issues batch:', error);
+                  throw error; // Re-throw to be caught by main execution
+                }
+              }
+            }
+
+            async function checkAndLabel(issue, owner, repo) {
+              if (!issue || !issue.parent) return;
+
+              let currentParent = issue.parent;
+              let tracedParents = [];
+              let matched = false;
+
+              while (currentParent) {
+                tracedParents.push(currentParent.url);
+
+                if (allowedParentUrls.includes(currentParent.url)) {
+                  console.log(`SUCCESS: Issue #${issue.number} is a descendant of ${currentParent.url}. Trace: ${tracedParents.join(' -> ')}. Adding label.`);
+                  await github.rest.issues.addLabels({
+                    owner,
+                    repo,
+                    issue_number: issue.number,
+                    labels: [labelToAdd]
+                  });
+                  matched = true;
+                  break;
+                }
+                currentParent = currentParent.parent;
+              }
+
+              if (!matched && context.eventName === 'issues') {
+                 console.log(`Issue #${issue.number} did not match any allowed workstreams. Trace: ${tracedParents.join(' -> ') || 'None'}.`);
+              }
+            }
+
+            // Main execution
+            try {
+              if (context.eventName === 'issues') {
+                console.log(`Processing single issue #${context.payload.issue.number}...`);
+                await processSingleIssue(context.repo.owner, context.repo.repo, context.payload.issue.number);
+              } else {
+                console.log(`Running for event: ${context.eventName}. Processing all open issues...`);
+                await processAllOpenIssues(context.repo.owner, context.repo.repo);
+              }
+            } catch (error) {
+              core.setFailed(`Workflow failed: ${error.message}`);
+            }

.github/workflows/pr-contribution-guidelines-notifier.yml

@@ -0,0 +1,83 @@
+name: '🏷️ PR Contribution Guidelines Notifier'
+
+on:
+  pull_request:
+    types:
+      - 'opened'
+
+jobs:
+  notify-process-change:
+    runs-on: 'ubuntu-latest'
+    if: |-
+      github.repository == 'google-gemini/gemini-cli' || github.repository == 'google-gemini/maintainers-gemini-cli'
+    permissions:
+      pull-requests: 'write'
+    steps:
+      - name: 'Generate GitHub App Token'
+        id: 'generate_token'
+        env:
+          APP_ID: '${{ secrets.APP_ID }}'
+        if: |-
+          ${{ env.APP_ID != '' }}
+        uses: 'actions/create-github-app-token@a8d616148505b5069dccd32f177bb87d7f39123b' # ratchet:actions/create-github-app-token@v2
+        with:
+          app-id: '${{ secrets.APP_ID }}'
+          private-key: '${{ secrets.PRIVATE_KEY }}'
+
+      - name: 'Check membership and post comment'
+        uses: 'actions/github-script@60a0d83039c74a4aee543508d2ffcb1c3799cdea'
+        with:
+          github-token: '${{ steps.generate_token.outputs.token || secrets.GITHUB_TOKEN }}'
+          script: |-
+            const org = context.repo.owner;
+            const repo = context.repo.repo;
+            const username = context.payload.pull_request.user.login;
+            const pr_number = context.payload.pull_request.number;
+
+            // 1. Check if the PR author is a maintainer
+            const authorAssociation = context.payload.pull_request.author_association;
+            if (['OWNER', 'MEMBER', 'COLLABORATOR'].includes(authorAssociation)) {
+              core.info(`${username} is a maintainer (Association: ${authorAssociation}). No notification needed.`);
+              return;
+            }
+
+            // 2. Check if the PR is already associated with an issue
+            const query = `
+              query($owner:String!, $repo:String!, $number:Int!) {
+                repository(owner:$owner, name:$repo) {
+                  pullRequest(number:$number) {
+                    closingIssuesReferences(first: 1) {
+                      totalCount
+                    }
+                  }
+                }
+              }
+            `;
+            const variables = { owner: org, repo: repo, number: pr_number };
+            const result = await github.graphql(query, variables);
+            const issueCount = result.repository.pullRequest.closingIssuesReferences.totalCount;
+
+            if (issueCount > 0) {
+              core.info(`PR #${pr_number} is already associated with an issue. No notification needed.`);
+              return;
+            }
+
+            // 3. Post the notification comment
+            core.info(`${username} is not a maintainer and PR #${pr_number} has no linked issue. Posting notification.`);
+
+            const comment = `
+            Hi @${username}, thank you so much for your contribution to Gemini CLI! We really appreciate the time and effort you've put into this.
+
+            We're making some updates to our contribution process to improve how we track and review changes. Please take a moment to review our recent discussion post: [Improving Our Contribution Process & Introducing New Guidelines](https://github.com/google-gemini/gemini-cli/discussions/16706).
+
+            Key Update: Starting **January 26, 2026**, the Gemini CLI project will require all pull requests to be associated with an existing issue. Any pull requests not linked to an issue by that date will be automatically closed.
+
+            Thank you for your understanding and for being a part of our community!
+            `.trim().replace(/^[ ]+/gm, '');
+
+            await github.rest.issues.createComment({
+              owner: org,
+              repo: repo,
+              issue_number: pr_number,
+              body: comment
+            });

.github/workflows/stale.yml

@@ -40,5 +40,5 @@ jobs:
             If this is still relevant, you are welcome to reopen or leave a comment. Thanks for contributing!
           days-before-stale: 60
           days-before-close: 14
-          exempt-issue-labels: 'pinned,security'
-          exempt-pr-labels: 'pinned,security'
+          exempt-issue-labels: 'pinned,security,🔒 maintainer only,help wanted,🗓️ Public Roadmap'
+          exempt-pr-labels: 'pinned,security,🔒 maintainer only,help wanted,🗓️ Public Roadmap'

.gitignore

@@ -11,6 +11,8 @@
 .gemini/*
 !.gemini/config.yaml
 !.gemini/commands/
+!.gemini/skills/
+!.gemini/settings.json
 
 # Note: .gemini-clipboard/ is NOT in gitignore so Gemini can access pasted images
 
@@ -53,7 +55,9 @@ gha-creds-*.json
 
 # Log files
 patch_output.log
+gemini-debug.log
 
 .genkit
 .gemini-clipboard/
 .eslintcache
+evals/logs/

.prettierignore

@@ -20,3 +20,4 @@ junit.xml
 .gemini-linters/
 Thumbs.db
 .pytest_cache
+**/SKILL.md

CONTRIBUTING.md

@@ -42,8 +42,13 @@ This project follows
 The process for contributing code is as follows:
 
 1.  **Find an issue** that you want to work on. If an issue is tagged as
-    "🔒Maintainers only", this means it is reserved for project maintainers. We
-    will not accept pull requests related to these issues.
+    `🔒Maintainers only`, this means it is reserved for project maintainers. We
+    will not accept pull requests related to these issues. In the near future,
+    we will explicitly mark issues looking for contributions using the
+    `help  wanted` label. If you believe an issue is a good candidate for
+    community contribution, please leave a comment on the issue. A maintainer
+    will review it and apply the `help-wanted` label if appropriate. Only
+    maintainers should attempt to add the `help-wanted` label to an issue.
 2.  **Fork the repository** and create a new branch.
 3.  **Make your changes** in the `packages/` directory.
 4.  **Ensure all checks pass** by running `npm run preflight`.
@@ -94,8 +99,11 @@ any code is written.
 - **For features:** The PR should be linked to the feature request or proposal
   issue that has been approved by a maintainer.
 
-If an issue for your change doesn't exist, please **open one first** and wait
-for feedback before you start coding.
+If an issue for your change doesn't exist, we will automatically close your PR
+along with a comment reminding you to associate the PR with an issue. The ideal
+workflow starts with an issue that has been reviewed and approved by a
+maintainer. Please **open the issue first** and wait for feedback before you
+start coding.
 
 #### 2. Keep it small and focused
 

GEMINI.md

@@ -1,409 +1,78 @@
-## Building and running
-
-Before submitting any changes, it is crucial to validate them by running the
-full preflight check. This command will build the repository, run all tests,
-check for type errors, and lint the code.
-
-To run the full suite of checks, execute the following command:
-
-```bash
-npm run preflight
-```
-
-This single command ensures that your changes meet all the quality gates of the
-project. While you can run the individual steps (`build`, `test`, `typecheck`,
-`lint`) separately, it is highly recommended to use `npm run preflight` to
-ensure a comprehensive validation.
-
-## Writing Tests
-
-This project uses **Vitest** as its primary testing framework. When writing
-tests, aim to follow existing patterns. Key conventions include:
-
-### Test Structure and Framework
-
-- **Framework**: All tests are written using Vitest (`describe`, `it`, `expect`,
-  `vi`).
-- **File Location**: Test files (`*.test.ts` for logic, `*.test.tsx` for React
-  components) are co-located with the source files they test.
-- **Configuration**: Test environments are defined in `vitest.config.ts` files.
-- **Setup/Teardown**: Use `beforeEach` and `afterEach`. Commonly,
-  `vi.resetAllMocks()` is called in `beforeEach` and `vi.restoreAllMocks()` in
-  `afterEach`.
-
-### Mocking (`vi` from Vitest)
-
-- **ES Modules**: Mock with
-  `vi.mock('module-name', async (importOriginal) => { ... })`. Use
-  `importOriginal` for selective mocking.
-  - _Example_:
-    `vi.mock('os', async (importOriginal) => { const actual = await importOriginal(); return { ...actual, homedir: vi.fn() }; });`
-- **Mocking Order**: For critical dependencies (e.g., `os`, `fs`) that affect
-  module-level constants, place `vi.mock` at the _very top_ of the test file,
-  before other imports.
-- **Hoisting**: Use `const myMock = vi.hoisted(() => vi.fn());` if a mock
-  function needs to be defined before its use in a `vi.mock` factory.
-- **Mock Functions**: Create with `vi.fn()`. Define behavior with
-  `mockImplementation()`, `mockResolvedValue()`, or `mockRejectedValue()`.
-- **Spying**: Use `vi.spyOn(object, 'methodName')`. Restore spies with
-  `mockRestore()` in `afterEach`.
-
-### Commonly Mocked Modules
-
-- **Node.js built-ins**: `fs`, `fs/promises`, `os` (especially `os.homedir()`),
-  `path`, `child_process` (`execSync`, `spawn`).
-- **External SDKs**: `@google/genai`, `@modelcontextprotocol/sdk`.
-- **Internal Project Modules**: Dependencies from other project packages are
-  often mocked.
-
-### React Component Testing (CLI UI - Ink)
-
-- Use `render()` from `ink-testing-library`.
-- Assert output with `lastFrame()`.
-- Wrap components in necessary `Context.Provider`s.
-- Mock custom React hooks and complex child components using `vi.mock()`.
-
-### Asynchronous Testing
-
-- Use `async/await`.
-- For timers, use `vi.useFakeTimers()`, `vi.advanceTimersByTimeAsync()`,
-  `vi.runAllTimersAsync()`.
-- Test promise rejections with `await expect(promise).rejects.toThrow(...)`.
-
-### General Guidance
-
-- When adding tests, first examine existing tests to understand and conform to
-  established conventions.
-- Pay close attention to the mocks at the top of existing test files; they
-  reveal critical dependencies and how they are managed in a test environment.
-
-## Git Repo
-
-The main branch for this project is called "main"
-
-## JavaScript/TypeScript
-
-When contributing to this React, Node, and TypeScript codebase, please
-prioritize the use of plain JavaScript objects with accompanying TypeScript
-interface or type declarations over JavaScript class syntax. This approach
-offers significant advantages, especially concerning interoperability with React
-and overall code maintainability.
-
-### Preferring Plain Objects over Classes
-
-JavaScript classes, by their nature, are designed to encapsulate internal state
-and behavior. While this can be useful in some object-oriented paradigms, it
-often introduces unnecessary complexity and friction when working with React's
-component-based architecture. Here's why plain objects are preferred:
-
-- Seamless React Integration: React components thrive on explicit props and
-  state management. Classes' tendency to store internal state directly within
-  instances can make prop and state propagation harder to reason about and
-  maintain. Plain objects, on the other hand, are inherently immutable (when
-  used thoughtfully) and can be easily passed as props, simplifying data flow
-  and reducing unexpected side effects.
-
-- Reduced Boilerplate and Increased Conciseness: Classes often promote the use
-  of constructors, this binding, getters, setters, and other boilerplate that
-  can unnecessarily bloat code. TypeScript interface and type declarations
-  provide powerful static type checking without the runtime overhead or
-  verbosity of class definitions. This allows for more succinct and readable
-  code, aligning with JavaScript's strengths in functional programming.
-
-- Enhanced Readability and Predictability: Plain objects, especially when their
-  structure is clearly defined by TypeScript interfaces, are often easier to
-  read and understand. Their properties are directly accessible, and there's no
-  hidden internal state or complex inheritance chains to navigate. This
-  predictability leads to fewer bugs and a more maintainable codebase.
-
-- Simplified Immutability: While not strictly enforced, plain objects encourage
-  an immutable approach to data. When you need to modify an object, you
-  typically create a new one with the desired changes, rather than mutating the
-  original. This pattern aligns perfectly with React's reconciliation process
-  and helps prevent subtle bugs related to shared mutable state.
-
-- Better Serialization and Deserialization: Plain JavaScript objects are
-  naturally easy to serialize to JSON and deserialize back, which is a common
-  requirement in web development (e.g., for API communication or local storage).
-  Classes, with their methods and prototypes, can complicate this process.
-
-### Embracing ES Module Syntax for Encapsulation
-
-Rather than relying on Java-esque private or public class members, which can be
-verbose and sometimes limit flexibility, we strongly prefer leveraging ES module
-syntax (`import`/`export`) for encapsulating private and public APIs.
-
-- Clearer Public API Definition: With ES modules, anything that is exported is
-  part of the public API of that module, while anything not exported is
-  inherently private to that module. This provides a very clear and explicit way
-  to define what parts of your code are meant to be consumed by other modules.
-
-- Enhanced Testability (Without Exposing Internals): By default, unexported
-  functions or variables are not accessible from outside the module. This
-  encourages you to test the public API of your modules, rather than their
-  internal implementation details. If you find yourself needing to spy on or
-  stub an unexported function for testing purposes, it's often a "code smell"
-  indicating that the function might be a good candidate for extraction into its
-  own separate, testable module with a well-defined public API. This promotes a
-  more robust and maintainable testing strategy.
-
-- Reduced Coupling: Explicitly defined module boundaries through import/export
-  help reduce coupling between different parts of your codebase. This makes it
-  easier to refactor, debug, and understand individual components in isolation.
-
-### Avoiding `any` Types and Type Assertions; Preferring `unknown`
-
-TypeScript's power lies in its ability to provide static type checking, catching
-potential errors before your code runs. To fully leverage this, it's crucial to
-avoid the `any` type and be judicious with type assertions.
-
-- **The Dangers of `any`**: Using any effectively opts out of TypeScript's type
-  checking for that particular variable or expression. While it might seem
-  convenient in the short term, it introduces significant risks:
-  - **Loss of Type Safety**: You lose all the benefits of type checking, making
-    it easy to introduce runtime errors that TypeScript would otherwise have
-    caught.
-  - **Reduced Readability and Maintainability**: Code with `any` types is harder
-    to understand and maintain, as the expected type of data is no longer
-    explicitly defined.
-  - **Masking Underlying Issues**: Often, the need for any indicates a deeper
-    problem in the design of your code or the way you're interacting with
-    external libraries. It's a sign that you might need to refine your types or
-    refactor your code.
-
-- **Preferring `unknown` over `any`**: When you absolutely cannot determine the
-  type of a value at compile time, and you're tempted to reach for any, consider
-  using unknown instead. unknown is a type-safe counterpart to any. While a
-  variable of type unknown can hold any value, you must perform type narrowing
-  (e.g., using typeof or instanceof checks, or a type assertion) before you can
-  perform any operations on it. This forces you to handle the unknown type
-  explicitly, preventing accidental runtime errors.
-
-  ```ts
-  function processValue(value: unknown) {
-    if (typeof value === 'string') {
-      // value is now safely a string
-      console.log(value.toUpperCase());
-    } else if (typeof value === 'number') {
-      // value is now safely a number
-      console.log(value * 2);
-    }
-    // Without narrowing, you cannot access properties or methods on 'value'
-    // console.log(value.someProperty); // Error: Object is of type 'unknown'.
-  }
-  ```
-
-- **Type Assertions (`as Type`) - Use with Caution**: Type assertions tell the
-  TypeScript compiler, "Trust me, I know what I'm doing; this is definitely of
-  this type." While there are legitimate use cases (e.g., when dealing with
-  external libraries that don't have perfect type definitions, or when you have
-  more information than the compiler), they should be used sparingly and with
-  extreme caution.
-  - **Bypassing Type Checking**: Like `any`, type assertions bypass TypeScript's
-    safety checks. If your assertion is incorrect, you introduce a runtime error
-    that TypeScript would not have warned you about.
-  - **Code Smell in Testing**: A common scenario where `any` or type assertions
-    might be tempting is when trying to test "private" implementation details
-    (e.g., spying on or stubbing an unexported function within a module). This
-    is a strong indication of a "code smell" in your testing strategy and
-    potentially your code structure. Instead of trying to force access to
-    private internals, consider whether those internal details should be
-    refactored into a separate module with a well-defined public API. This makes
-    them inherently testable without compromising encapsulation.
-
-### Type narrowing `switch` clauses
-
-Use the `checkExhaustive` helper in the default clause of a switch statement.
-This will ensure that all of the possible options within the value or
-enumeration are used.
-
-This helper method can be found in `packages/cli/src/utils/checks.ts`
-
-### Embracing JavaScript's Array Operators
-
-To further enhance code cleanliness and promote safe functional programming
-practices, leverage JavaScript's rich set of array operators as much as
-possible. Methods like `.map()`, `.filter()`, `.reduce()`, `.slice()`,
-`.sort()`, and others are incredibly powerful for transforming and manipulating
-data collections in an immutable and declarative way.
-
-Using these operators:
-
-- Promotes Immutability: Most array operators return new arrays, leaving the
-  original array untouched. This functional approach helps prevent unintended
-  side effects and makes your code more predictable.
-- Improves Readability: Chaining array operators often lead to more concise and
-  expressive code than traditional for loops or imperative logic. The intent of
-  the operation is clear at a glance.
-- Facilitates Functional Programming: These operators are cornerstones of
-  functional programming, encouraging the creation of pure functions that take
-  inputs and produce outputs without causing side effects. This paradigm is
-  highly beneficial for writing robust and testable code that pairs well with
-  React.
-
-By consistently applying these principles, we can maintain a codebase that is
-not only efficient and performant but also a joy to work with, both now and in
-the future.
-
-## React (mirrored and adjusted from [react-mcp-server](https://github.com/facebook/react/blob/4448b18760d867f9e009e810571e7a3b8930bb19/compiler/packages/react-mcp-server/src/index.ts#L376C1-L441C94))
-
-### Role
-
-You are a React assistant that helps users write more efficient and optimizable
-React code. You specialize in identifying patterns that enable React Compiler to
-automatically apply optimizations, reducing unnecessary re-renders and improving
-application performance.
-
-### Follow these guidelines in all code you produce and suggest
-
-Use functional components with Hooks: Do not generate class components or use
-old lifecycle methods. Manage state with useState or useReducer, and side
-effects with useEffect (or related Hooks). Always prefer functions and Hooks for
-any new component logic.
-
-Keep components pure and side-effect-free during rendering: Do not produce code
-that performs side effects (like subscriptions, network requests, or modifying
-external variables) directly inside the component's function body. Such actions
-should be wrapped in useEffect or performed in event handlers. Ensure your
-render logic is a pure function of props and state.
-
-Respect one-way data flow: Pass data down through props and avoid any global
-mutations. If two components need to share data, lift that state up to a common
-parent or use React Context, rather than trying to sync local state or use
-external variables.
-
-Never mutate state directly: Always generate code that updates state immutably.
-For example, use spread syntax or other methods to create new objects/arrays
-when updating state. Do not use assignments like state.someValue = ... or array
-mutations like array.push() on state variables. Use the state setter (setState
-from useState, etc.) to update state.
-
-Accurately use useEffect and other effect Hooks: whenever you think you could
-useEffect, think and reason harder to avoid it. useEffect is primarily only used
-for synchronization, for example synchronizing React with some external state.
-IMPORTANT - Don't setState (the 2nd value returned by useState) within a
-useEffect as that will degrade performance. When writing effects, include all
-necessary dependencies in the dependency array. Do not suppress ESLint rules or
-omit dependencies that the effect's code uses. Structure the effect callbacks to
-handle changing values properly (e.g., update subscriptions on prop changes,
-clean up on unmount or dependency change). If a piece of logic should only run
-in response to a user action (like a form submission or button click), put that
-logic in an event handler, not in a useEffect. Where possible, useEffects should
-return a cleanup function.
-
-Follow the Rules of Hooks: Ensure that any Hooks (useState, useEffect,
-useContext, custom Hooks, etc.) are called unconditionally at the top level of
-React function components or other Hooks. Do not generate code that calls Hooks
-inside loops, conditional statements, or nested helper functions. Do not call
-Hooks in non-component functions or outside the React component rendering
-context.
-
-Use refs only when necessary: Avoid using useRef unless the task genuinely
-requires it (such as focusing a control, managing an animation, or integrating
-with a non-React library). Do not use refs to store application state that
-should be reactive. If you do use refs, never write to or read from ref.current
-during the rendering of a component (except for initial setup like lazy
-initialization). Any ref usage should not affect the rendered output directly.
-
-Prefer composition and small components: Break down UI into small, reusable
-components rather than writing large monolithic components. The code you
-generate should promote clarity and reusability by composing components
-together. Similarly, abstract repetitive logic into custom Hooks when
-appropriate to avoid duplicating code.
-
-Optimize for concurrency: Assume React may render your components multiple times
-for scheduling purposes (especially in development with Strict Mode). Write code
-that remains correct even if the component function runs more than once. For
-instance, avoid side effects in the component body and use functional state
-updates (e.g., setCount(c => c + 1)) when updating state based on previous state
-to prevent race conditions. Always include cleanup functions in effects that
-subscribe to external resources. Don't write useEffects for "do this when this
-changes" side effects. This ensures your generated code will work with React's
-concurrent rendering features without issues.
-
-Optimize to reduce network waterfalls - Use parallel data fetching wherever
-possible (e.g., start multiple requests at once rather than one after another).
-Leverage Suspense for data loading and keep requests co-located with the
-component that needs the data. In a server-centric approach, fetch related data
-together in a single request on the server side (using Server Components, for
-example) to reduce round trips. Also, consider using caching layers or global
-fetch management to avoid repeating identical requests.
-
-Rely on React Compiler - useMemo, useCallback, and React.memo can be omitted if
-React Compiler is enabled. Avoid premature optimization with manual memoization.
-Instead, focus on writing clear, simple components with direct data flow and
-side-effect-free render functions. Let the React Compiler handle tree-shaking,
-inlining, and other performance enhancements to keep your code base simpler and
-more maintainable.
-
-Design for a good user experience - Provide clear, minimal, and non-blocking UI
-states. When data is loading, show lightweight placeholders (e.g., skeleton
-screens) rather than intrusive spinners everywhere. Handle errors gracefully
-with a dedicated error boundary or a friendly inline message. Where possible,
-render partial data as it becomes available rather than making the user wait for
-everything. Suspense allows you to declare the loading states in your component
-tree in a natural way, preventing “flash” states and improving perceived
-performance.
-
-### Process
-
-1. Analyze the user's code for optimization opportunities:
-   - Check for React anti-patterns that prevent compiler optimization
-   - Look for component structure issues that limit compiler effectiveness
-   - Think about each suggestion you are making and consult React docs for best
-     practices
-
-2. Provide actionable guidance:
-   - Explain specific code changes with clear reasoning
-   - Show before/after examples when suggesting changes
-   - Only suggest changes that meaningfully improve optimization potential
-
-### Optimization Guidelines
-
-- State updates should be structured to enable granular updates
-- Side effects should be isolated and dependencies clearly defined
-
-## Documentation guidelines
-
-When working in the `/docs` directory, follow the guidelines in this section:
-
-- **Role:** You are an expert technical writer and AI assistant for contributors
-  to Gemini CLI. Produce professional, accurate, and consistent documentation to
-  guide users of Gemini CLI.
-- **Technical Accuracy:** Do not invent facts, commands, code, API names, or
-  output. All technical information specific to Gemini CLI must be based on code
-  found within this directory and its subdirectories.
-- **Style Authority:** Your source for writing guidance and style is the
-  "Documentation contribution process" section in the root directory's
-  `CONTRIBUTING.md` file, as well as any guidelines provided this section.
-- **Information Architecture Consideration:** Before proposing documentation
-  changes, consider the information architecture. If a change adds significant
-  new content to existing documents, evaluate if creating a new, more focused
-  page or changes to `sidebar.json` would provide a better user experience.
-- **Proactive User Consideration:** The user experience should be a primary
-  concern when making changes to documentation. Aim to fill gaps in existing
-  knowledge whenever possible while keeping documentation concise and easy for
-  users to understand. If changes might hinder user understanding or
-  accessibility, proactively raise these concerns and propose alternatives.
-
-## Comments policy
-
-Only write high-value comments if at all. Avoid talking to the user through
-comments.
-
-## Logging and Error Handling
-
-- **Avoid Console Statements:** Do not use `console.log`, `console.error`, or
-  similar methods directly.
-- **Non-User-Facing Logs:** For developer-facing debug messages, use
-  `debugLogger` (from `@google/gemini-cli-core`).
-- **User-Facing Feedback:** To surface errors or warnings to the user, use
-  `coreEvents.emitFeedback` (from `@google/gemini-cli-core`).
-
-## General requirements
-
-- If there is something you do not understand or is ambiguous, seek confirmation
-  or clarification from the user before making changes based on assumptions.
-- Use hyphens instead of underscores in flag names (e.g. `my-flag` instead of
-  `my_flag`).
-- Always refer to Gemini CLI as `Gemini CLI`, never `the Gemini CLI`.
+# Gemini CLI Project Context
+
+Gemini CLI is an open-source AI agent that brings the power of Gemini directly
+into the terminal. It is designed to be a terminal-first, extensible, and
+powerful tool for developers.
+
+## Project Overview
+
+- **Purpose:** Provide a seamless terminal interface for Gemini models,
+  supporting code understanding, generation, automation, and integration via MCP
+  (Model Context Protocol).
+- **Main Technologies:**
+  - **Runtime:** Node.js (>=20.0.0, recommended ~20.19.0 for development)
+  - **Language:** TypeScript
+  - **UI Framework:** React (using [Ink](https://github.com/vadimdemedes/ink)
+    for CLI rendering)
+  - **Testing:** Vitest
+  - **Bundling:** esbuild
+  - **Linting/Formatting:** ESLint, Prettier
+- **Architecture:** Monorepo structure using npm workspaces.
+  - `packages/cli`: User-facing terminal UI, input processing, and display
+    rendering.
+  - `packages/core`: Backend logic, Gemini API orchestration, prompt
+    construction, and tool execution.
+  - `packages/core/src/tools/`: Built-in tools for file system, shell, and web
+    operations.
+  - `packages/a2a-server`: Experimental Agent-to-Agent server.
+  - `packages/vscode-ide-companion`: VS Code extension pairing with the CLI.
+
+## Building and Running
+
+- **Install Dependencies:** `npm install`
+- **Build All:** `npm run build:all` (Builds packages, sandbox, and VS Code
+  companion)
+- **Build Packages:** `npm run build`
+- **Run in Development:** `npm run start`
+- **Run in Debug Mode:** `npm run debug` (Enables Node.js inspector)
+- **Bundle Project:** `npm run bundle`
+- **Clean Artifacts:** `npm run clean`
+
+## Testing and Quality
+
+- **Test Commands:**
+  - **Unit (All):** `npm run test`
+  - **Integration (E2E):** `npm run test:e2e`
+  - **Workspace-Specific:** `npm test -w <pkg> -- <path>` (Note: `<path>` must
+    be relative to the workspace root, e.g.,
+    `-w @google/gemini-cli-core -- src/routing/modelRouterService.test.ts`)
+- **Full Validation:** `npm run preflight` (Heaviest check; runs clean, install,
+  build, lint, type check, and tests. Recommended before submitting PRs.)
+- **Individual Checks:** `npm run lint` / `npm run format` / `npm run typecheck`
+
+## Development Conventions
+
+- **Contributions:** Follow the process outlined in `CONTRIBUTING.md`. Requires
+  signing the Google CLA.
+- **Pull Requests:** Keep PRs small, focused, and linked to an existing issue.
+- **Commit Messages:** Follow the
+  [Conventional Commits](https://www.conventionalcommits.org/) standard.
+- **Coding Style:** Adhere to existing patterns in `packages/cli` (React/Ink)
+  and `packages/core` (Backend logic).
+- **Imports:** Use specific imports and avoid restricted relative imports
+  between packages (enforced by ESLint).
+
+## Testing Conventions
+
+- **Environment Variables:** When testing code that depends on environment
+  variables, use `vi.stubEnv('NAME', 'value')` in `beforeEach` and
+  `vi.unstubAllEnvs()` in `afterEach`. Avoid modifying `process.env` directly as
+  it can lead to test leakage and is less reliable. To "unset" a variable, use
+  an empty string `vi.stubEnv('NAME', '')`.
+
+## Documentation
+
+- Suggest documentation updates when code changes render existing documentation
+  obsolete or incomplete.
+- Located in the `docs/` directory.
+- Use the `docs-writer` skill.

README.md

@@ -4,7 +4,7 @@
 [![Gemini CLI E2E (Chained)](https://github.com/google-gemini/gemini-cli/actions/workflows/chained_e2e.yml/badge.svg)](https://github.com/google-gemini/gemini-cli/actions/workflows/chained_e2e.yml)
 [![Version](https://img.shields.io/npm/v/@google/gemini-cli)](https://www.npmjs.com/package/@google/gemini-cli)
 [![License](https://img.shields.io/github/license/google-gemini/gemini-cli)](https://github.com/google-gemini/gemini-cli/blob/main/LICENSE)
-[![View Code Wiki](https://www.gstatic.com/_/boq-sdlc-agents-ui/_/r/YUi5dj2UWvE.svg)](https://codewiki.google/github.com/google-gemini/gemini-cli)
+[![View Code Wiki](https://assets.codewiki.google/readme-badge/static.svg)](https://codewiki.google/github.com/google-gemini/gemini-cli?utm_source=badge&utm_medium=github&utm_campaign=github.com/google-gemini/gemini-cli)
 
 ![Gemini CLI Screenshot](./docs/assets/gemini-screenshot.png)
 
@@ -18,7 +18,8 @@ Learn all about Gemini CLI in our [documentation](https://geminicli.com/docs/).
 
 - **🎯 Free tier**: 60 requests/min and 1,000 requests/day with personal Google
   account.
-- **🧠 Powerful Gemini 2.5 Pro**: Access to 1M token context window.
+- **🧠 Powerful Gemini 3 models**: Access to improved reasoning and 1M token
+  context window.
 - **🔧 Built-in tools**: Google Search grounding, file operations, shell
   commands, web fetching.
 - **🔌 Extensible**: MCP (Model Context Protocol) support for custom
@@ -54,6 +55,23 @@ npm install -g @google/gemini-cli
 brew install gemini-cli
 ```
 
+#### Install globally with MacPorts (macOS)
+
+```bash
+sudo port install gemini-cli
+```
+
+#### Install with Anaconda (for restricted environments)
+
+```bash
+# Create and activate a new environment
+conda create -y -n gemini_env -c conda-forge nodejs
+conda activate gemini_env
+
+# Install Gemini CLI globally via npm (inside the environment)
+npm install -g @google/gemini-cli
+```
+
 ## Release Cadence and Tags
 
 See [Releases](./docs/releases.md) for more details.
@@ -140,7 +158,7 @@ for details)
 **Benefits:**
 
 - **Free tier**: 60 requests/min and 1,000 requests/day
-- **Gemini 2.5 Pro** with 1M token context window
+- **Gemini 3 models** with 1M token context window
 - **No API key management** - just sign in with your Google account
 - **Automatic updates** to latest models
 
@@ -164,7 +182,7 @@ gemini
 
 **Benefits:**
 
-- **Free tier**: 100 requests/day with Gemini 2.5 Pro
+- **Free tier**: 1000 requests/day with Gemini 3 (mix of flash and pro)
 - **Model selection**: Choose specific Gemini models
 - **Usage-based billing**: Upgrade for higher limits when needed
 
@@ -371,5 +389,5 @@ See the [Uninstall Guide](docs/cli/uninstall.md) for removal instructions.
 ---
 
 <p align="center">
-  Built with ❤️ by Google and the open source community
+  Built with ❤️ by Google, shreya, and the open source community
 </p>

docs/architecture.md

@@ -13,11 +13,11 @@ input:
       as handling the initial user input, presenting the final output, and
       managing the overall user experience.
     - **Key functions contained in the package:**
-      - [Input processing](/docs/cli/commands.md)
+      - [Input processing](/docs/cli/commands)
       - History management
       - Display rendering
-      - [Theme and UI customization](/docs/cli/themes.md)
-      - [CLI configuration settings](/docs/get-started/configuration.md)
+      - [Theme and UI customization](/docs/cli/themes)
+      - [CLI configuration settings](/docs/get-started/configuration)
 
 2.  **Core package (`packages/core`):**
     - **Purpose:** This acts as the backend for the Gemini CLI. It receives

docs/changelogs/index.md

@@ -18,6 +18,118 @@ on GitHub.
 | [Preview](preview.md) | Experimental features ready for early feedback. |
 | [Stable](latest.md)   | Stable, recommended for general use.            |
 
+## Announcements: v0.25.0 - 2026-01-20
+
+- **Skills and Agents Improvements:** We've enhanced the `activate_skill` tool,
+  added a new `pr-creator` skill
+  ([#16232](https://github.com/google-gemini/gemini-cli/pull/16232) by
+  [@NTaylorMullen](https://github.com/NTaylorMullen)), enabled skills by
+  default, improved the `cli_help` agent
+  ([#16100](https://github.com/google-gemini/gemini-cli/pull/16100) by
+  [@scidomino](https://github.com/scidomino)), and added a new `/agents refresh`
+  command ([#16204](https://github.com/google-gemini/gemini-cli/pull/16204) by
+  [@joshualitt](https://github.com/joshualitt)).
+- **UI/UX Refinements:** You'll notice more transparent feedback for skills
+  ([#15954](https://github.com/google-gemini/gemini-cli/pull/15954) by
+  [@NTaylorMullen](https://github.com/NTaylorMullen)), the ability to switch
+  focus between the shell and input with Tab
+  ([#14332](https://github.com/google-gemini/gemini-cli/pull/14332) by
+  [@jacob314](https://github.com/jacob314)), and dynamic terminal tab titles
+  ([#16378](https://github.com/google-gemini/gemini-cli/pull/16378) by
+  [@NTaylorMullen](https://github.com/NTaylorMullen)).
+- **Core Functionality & Performance:** This release includes support for
+  built-in agent skills
+  ([#16045](https://github.com/google-gemini/gemini-cli/pull/16045) by
+  [@NTaylorMullen](https://github.com/NTaylorMullen)), refined Gemini 3 system
+  instructions ([#16139](https://github.com/google-gemini/gemini-cli/pull/16139)
+  by [@NTaylorMullen](https://github.com/NTaylorMullen)), caching for ignore
+  instances to improve performance
+  ([#16185](https://github.com/google-gemini/gemini-cli/pull/16185) by
+  [@EricRahm](https://github.com/EricRahm)), and enhanced retry mechanisms
+  ([#16489](https://github.com/google-gemini/gemini-cli/pull/16489) by
+  [@sehoon38](https://github.com/sehoon38)).
+- **Bug Fixes and Stability:** We've squashed numerous bugs across the CLI,
+  core, and workflows, addressing issues with subagent delegation, unicode
+  character crashes, and sticky header regressions.
+
+## Announcements: v0.24.0 - 2026-01-14
+
+- **Agent Skills:** We've introduced significant advancements in Agent Skills.
+  This includes initial documentation and tutorials to help you get started,
+  alongside enhanced support for remote agents, allowing for more distributed
+  and powerful automation within Gemini CLI.
+  ([#15869](https://github.com/google-gemini/gemini-cli/pull/15869) by
+  [@NTaylorMullen](https://github.com/NTaylorMullen)),
+  ([#16013](https://github.com/google-gemini/gemini-cli/pull/16013) by
+  [@adamweidman](https://github.com/adamweidman))
+- **Improved UI/UX:** The user interface has received several updates, featuring
+  visual indicators for hook execution, a more refined display for settings, and
+  the ability to use the Tab key to effortlessly switch focus between the shell
+  and input areas.
+  ([#15408](https://github.com/google-gemini/gemini-cli/pull/15408) by
+  [@abhipatel12](https://github.com/abhipatel12)),
+  ([#14332](https://github.com/google-gemini/gemini-cli/pull/14332) by
+  [@galz10](https://github.com/galz10))
+- **Enhanced Security:** Security has been a major focus, with default folder
+  trust now set to untrusted for increased safety. The Policy Engine has been
+  improved to allow specific modes in user and administrator policies, and
+  granular allowlisting for shell commands has been implemented, providing finer
+  control over tool execution.
+  ([#15943](https://github.com/google-gemini/gemini-cli/pull/15943) by
+  [@galz10](https://github.com/galz10)),
+  ([#15977](https://github.com/google-gemini/gemini-cli/pull/15977) by
+  [@NTaylorMullen](https://github.com/NTaylorMullen))
+- **Core Functionality:** This release includes a mandatory MessageBus
+  injection, marking Phase 3 of a hard migration to a more robust internal
+  communication system. We've also added support for built-in skills with the
+  CLI itself, and enhanced model routing to effectively utilize subagents.
+  ([#15776](https://github.com/google-gemini/gemini-cli/pull/15776) by
+  [@abhipatel12](https://github.com/abhipatel12)),
+  ([#16300](https://github.com/google-gemini/gemini-cli/pull/16300) by
+  [@NTaylorMullen](https://github.com/NTaylorMullen))
+- **Terminal Features:** Terminal interactions are more seamless with new
+  features like OSC 52 paste support, along with fixes for Windows clipboard
+  paste issues and general improvements to pasting in Windows terminals.
+  ([#15336](https://github.com/google-gemini/gemini-cli/pull/15336) by
+  [@scidomino](https://github.com/scidomino)),
+  ([#15932](https://github.com/google-gemini/gemini-cli/pull/15932) by
+  [@scidomino](https://github.com/scidomino))
+- **New Commands:** To manage the new features, we've added several new
+  commands: `/agents refresh` to update agent configurations, `/skills reload`
+  to refresh skill definitions, and `/skills install/uninstall` for easier
+  management of your Agent Skills.
+  ([#16204](https://github.com/google-gemini/gemini-cli/pull/16204) by
+  [@NTaylorMullen](https://github.com/NTaylorMullen)),
+  ([#15865](https://github.com/google-gemini/gemini-cli/pull/15865) by
+  [@NTaylorMullen](https://github.com/NTaylorMullen)),
+  ([#16377](https://github.com/google-gemini/gemini-cli/pull/16377) by
+  [@NTaylorMullen](https://github.com/NTaylorMullen))
+
+## Announcements: v0.23.0 - 2026-01-07
+
+- 🎉 **Experimental Agent Skills Support in Preview:** Gemini CLI now supports
+  [Agent Skills](https://agentskills.io/home) in our preview builds. This is an
+  early preview where we’re looking for feedback!
+  - Install Preview: `npm install -g @google/gemini-cli@preview`
+  - Enable in `/settings`
+  - Docs:
+    [https://geminicli.com/docs/cli/skills/](https://geminicli.com/docs/cli/skills/)
+- **Gemini CLI wrapped:** Run `npx gemini-wrapped` to visualize your usage
+  stats, top models, languages, and more!
+- **Windows clipboard image support:** Windows users can now paste images
+  directly from their clipboard into the CLI using `Alt`+`V`.
+  ([pr](https://github.com/google-gemini/gemini-cli/pull/13997) by
+  [@sgeraldes](https://github.com/sgeraldes))
+- **Terminal background color detection:** Automatically optimizes your
+  terminal's background color to select compatible themes and provide
+  accessibility warnings.
+  ([pr](https://github.com/google-gemini/gemini-cli/pull/15132) by
+  [@jacob314](https://github.com/jacob314))
+- **Session logout:** Use the new `/logout` command to instantly clear
+  credentials and reset your authentication state for seamless account
+  switching. ([pr](https://github.com/google-gemini/gemini-cli/pull/13383) by
+  [@CN-Scars](https://github.com/CN-Scars))
+
 ## Announcements: v0.22.0 - 2025-12-22
 
 - 🎉**Free Tier + Gemini 3:** Free tier users now all have access to Gemini 3

docs/changelogs/latest.md

@@ -1,6 +1,6 @@
-# Latest stable release: v0.22.0 - v0.22.5
+# Latest stable release: v0.25.0
 
-Released: December 30, 2025
+Released: January 20, 2026
 
 For most users, our latest stable release is the recommended release. Install
 the latest stable version with:
@@ -11,143 +11,360 @@ npm install -g @google/gemini-cli
 
 ## Highlights
 
-- **Comprehensive quota visibility:** View usage statistics for all available
-  models in the `/stats` command, even those not yet used in your current
-  session. ([pic](https://imgur.com/a/cKyDtYh),
-  [pr](https://github.com/google-gemini/gemini-cli/pull/14764) by
-  [@sehoon38](https://github.com/sehoon38))
-- **Polished CLI statistics:** We’ve cleaned up the `/stats` view to prioritize
-  actionable quota information while providing a detailed token and
-  cache-efficiency breakdown in `/stats model`
-  ([login with Google](https://imgur.com/a/w9xKthm),
-  [api key](https://imgur.com/a/FjQPHOY),
-  [model stats](https://imgur.com/a/VfWzVgw),
-  [pr](https://github.com/google-gemini/gemini-cli/pull/14961) by
-  [@jacob314](https://github.com/jacob314))
-- **Multi-file drag & drop:** Multi-file drag & drop is now supported and
-  properly translated to be prefixed with `@`.
-  ([pr](https://github.com/google-gemini/gemini-cli/pull/14832) by
-  [@jackwotherspoon](https://github.com/jackwotherspoon))
+- **Skills and Agents Improvements:** Enhanced `activate_skill` tool, new
+  `pr-creator` skill, default enablement of skills, improved `cli_help` agent,
+  and a new `/agents refresh` command.
+- **UI/UX Refinements:** Transparent feedback for skills, ability to switch
+  focus between shell and input with Tab, and dynamic terminal tab titles.
+- **Core Functionality & Performance:** Support for built-in agent skills,
+  refined Gemini 3 system instructions, caching ignore instances for
+  performance, and improved retry mechanisms.
+- **Bug Fixes and Stability:** Numerous bug fixes across the CLI, core, and
+  workflows, including issues with subagent delegation, unicode character
+  crashes, and sticky header regressions.
 
 ## What's Changed
 
-- feat(ide): fallback to GEMINI_CLI_IDE_AUTH_TOKEN env var by @skeshive in
-  https://github.com/google-gemini/gemini-cli/pull/14843
-- feat: display quota stats for unused models in /stats by @sehoon38 in
-  https://github.com/google-gemini/gemini-cli/pull/14764
-- feat: ensure codebase investigator uses preview model when main agent does by
-  @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/14412
-- chore: add closing reason to stale bug workflow by @galz10 in
-  https://github.com/google-gemini/gemini-cli/pull/14861
-- Send the model and CLI version with the user agent by @gundermanc in
-  https://github.com/google-gemini/gemini-cli/pull/14865
-- refactor(sessions): move session summary generation to startup by
-  @jackwotherspoon in https://github.com/google-gemini/gemini-cli/pull/14691
-- Limit search depth in path corrector by @scidomino in
-  https://github.com/google-gemini/gemini-cli/pull/14869
-- Fix: Correct typo in code comment by @kuishou68 in
-  https://github.com/google-gemini/gemini-cli/pull/14671
-- feat(core): Plumbing for late resolution of model configs. by @joshualitt in
-  https://github.com/google-gemini/gemini-cli/pull/14597
-- feat: attempt more error parsing by @adamfweidman in
-  https://github.com/google-gemini/gemini-cli/pull/14899
-- Add missing await. by @gundermanc in
-  https://github.com/google-gemini/gemini-cli/pull/14910
-- feat(core): Add support for transcript_path in hooks for git-ai/Gemini
-  extension by @svarlamov in
-  https://github.com/google-gemini/gemini-cli/pull/14663
-- refactor: implement DelegateToAgentTool with discriminated union by
-  @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/14769
-- feat: reset availabilityService on /auth by @adamfweidman in
-  https://github.com/google-gemini/gemini-cli/pull/14911
-- chore/release: bump version to 0.21.0-nightly.20251211.8c83e1ea9 by
-  @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/14924
-- Fix: Correctly detect MCP tool errors by @kevin-ramdass in
-  https://github.com/google-gemini/gemini-cli/pull/14937
-- increase labeler timeout by @scidomino in
-  https://github.com/google-gemini/gemini-cli/pull/14922
-- tool(cli): tweak the frontend tool to be aware of more core files from the cli
-  by @jacob314 in https://github.com/google-gemini/gemini-cli/pull/14962
-- feat(cli): polish cached token stats and simplify stats display when quota is
-  present. by @jacob314 in
-  https://github.com/google-gemini/gemini-cli/pull/14961
-- feat(settings-validation): add validation for settings schema by @lifefloating
-  in https://github.com/google-gemini/gemini-cli/pull/12929
-- fix(ide): Update IDE extension to write auth token in env var by @skeshive in
-  https://github.com/google-gemini/gemini-cli/pull/14999
-- Revert "chore(deps): bump express from 5.1.0 to 5.2.0" by @skeshive in
-  https://github.com/google-gemini/gemini-cli/pull/14998
-- feat(a2a): Introduce /init command for a2a server by @cocosheng-g in
-  https://github.com/google-gemini/gemini-cli/pull/13419
-- feat: support multi-file drag and drop of images by @jackwotherspoon in
-  https://github.com/google-gemini/gemini-cli/pull/14832
-- fix(policy): allow codebase_investigator by default in read-only policy by
-  @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/15000
-- refactor(ide ext): Update port file name + switch to 1-based index for
-  characters + remove truncation text by @skeshive in
-  https://github.com/google-gemini/gemini-cli/pull/10501
-- fix(vscode-ide-companion): correct license generation for workspace
-  dependencies by @skeshive in
-  https://github.com/google-gemini/gemini-cli/pull/15004
-- fix: temp fix for subagent invocation until subagent delegation is merged to
-  stable by @abhipatel12 in
-  https://github.com/google-gemini/gemini-cli/pull/15007
-- test: update ide detection tests to make them more robust when run in an ide
-  by @kevin-ramdass in https://github.com/google-gemini/gemini-cli/pull/15008
-- Remove flex from stats display. See snapshots for diffs. by @jacob314 in
-  https://github.com/google-gemini/gemini-cli/pull/14983
-- Add license field into package.json by @jb-perez in
-  https://github.com/google-gemini/gemini-cli/pull/14473
-- feat: Persistent "Always Allow" policies with granular shell & MCP support by
-  @allenhutchison in https://github.com/google-gemini/gemini-cli/pull/14737
-- chore/release: bump version to 0.21.0-nightly.20251212.54de67536 by
-  @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/14969
-- fix(core): commandPrefix word boundary and compound command safety by
-  @allenhutchison in https://github.com/google-gemini/gemini-cli/pull/15006
-- chore(docs): add 'Maintainers only' label info to CONTRIBUTING.md by @jacob314
-  in https://github.com/google-gemini/gemini-cli/pull/14914
-- Refresh hooks when refreshing extensions. by @scidomino in
-  https://github.com/google-gemini/gemini-cli/pull/14918
-- Add clarity to error messages by @gsehgal in
-  https://github.com/google-gemini/gemini-cli/pull/14879
-- chore : remove a redundant tip by @JayadityaGit in
-  https://github.com/google-gemini/gemini-cli/pull/14947
-- chore/release: bump version to 0.21.0-nightly.20251213.977248e09 by
-  @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/15029
-- Disallow redundant typecasts. by @gundermanc in
-  https://github.com/google-gemini/gemini-cli/pull/15030
-- fix(auth): prioritize GEMINI_API_KEY env var and skip unnecessary key… by
-  @galz10 in https://github.com/google-gemini/gemini-cli/pull/14745
-- fix: use zod for safety check result validation by @allenhutchison in
-  https://github.com/google-gemini/gemini-cli/pull/15026
-- update(telemetry): add hashed_extension_name to field to extension events by
-  @kiranani in https://github.com/google-gemini/gemini-cli/pull/15025
-- fix: similar to policy-engine, throw error in case of requiring tool execution
-  confirmation for non-interactive mode by @MayV in
-  https://github.com/google-gemini/gemini-cli/pull/14702
-- Clean up processes in integration tests by @scidomino in
-  https://github.com/google-gemini/gemini-cli/pull/15102
-- docs: update policy engine getting started and defaults by @NTaylorMullen in
-  https://github.com/google-gemini/gemini-cli/pull/15105
-- Fix tool output fragmentation by encapsulating content in functionResponse by
-  @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/13082
-- Simplify method signature. by @scidomino in
-  https://github.com/google-gemini/gemini-cli/pull/15114
-- Show raw input token counts in json output. by @jacob314 in
-  https://github.com/google-gemini/gemini-cli/pull/15021
-- fix: Mark A2A requests as interactive by @MayV in
-  https://github.com/google-gemini/gemini-cli/pull/15108
-- use previewFeatures to determine which pro model to use for A2A by @sehoon38
-  in https://github.com/google-gemini/gemini-cli/pull/15131
-- refactor(cli): fix settings merging so that settings using the new json format
-  take priority over ones using the old format by @jacob314 in
-  https://github.com/google-gemini/gemini-cli/pull/15116
-- fix(patch): cherry-pick a6d1245 to release/v0.22.0-preview.1-pr-15214 to patch
-  version v0.22.0-preview.1 and create version 0.22.0-preview.2 by
-  @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/15226
-- fix(patch): cherry-pick 9e6914d to release/v0.22.0-preview.2-pr-15288 to patch
-  version v0.22.0-preview.2 and create version 0.22.0-preview.3 by
-  @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/15294
+- feat(core): improve activate_skill tool and use lowercase XML tags by
+  @NTaylorMullen in
+  [#16009](https://github.com/google-gemini/gemini-cli/pull/16009)
+- Add initiation method telemetry property by @gundermanc in
+  [#15818](https://github.com/google-gemini/gemini-cli/pull/15818)
+- chore(release): bump version to 0.25.0-nightly.20260107.59a18e710 by
+  @gemini-cli-robot in
+  [#16048](https://github.com/google-gemini/gemini-cli/pull/16048)
+- Hx support by @kevinfjiang in
+  [#16032](https://github.com/google-gemini/gemini-cli/pull/16032)
+- [Skills] Foundation: Centralize management logic and feedback rendering by
+  @NTaylorMullen in
+  [#15952](https://github.com/google-gemini/gemini-cli/pull/15952)
+- Introduce GEMINI_CLI_HOME for strict test isolation by @NTaylorMullen in
+  [#15907](https://github.com/google-gemini/gemini-cli/pull/15907)
+- [Skills] Multi-scope skill enablement and shadowing fix by @NTaylorMullen in
+  [#15953](https://github.com/google-gemini/gemini-cli/pull/15953)
+- policy: extract legacy policy from core tool scheduler to policy engine by
+  @abhipatel12 in
+  [#15902](https://github.com/google-gemini/gemini-cli/pull/15902)
+- Enhance TestRig with process management and timeouts by @NTaylorMullen in
+  [#15908](https://github.com/google-gemini/gemini-cli/pull/15908)
+- Update troubleshooting doc for UNABLE_TO_GET_ISSUER_CERT_LOCALLY by @sehoon38
+  in [#16069](https://github.com/google-gemini/gemini-cli/pull/16069)
+- Add keytar to dependencies by @chrstnb in
+  [#15928](https://github.com/google-gemini/gemini-cli/pull/15928)
+- Simplify extension settings command by @chrstnb in
+  [#16001](https://github.com/google-gemini/gemini-cli/pull/16001)
+- feat(admin): implement extensions disabled by @skeshive in
+  [#16024](https://github.com/google-gemini/gemini-cli/pull/16024)
+- Core data structure updates for Rewind functionality by @Adib234 in
+  [#15714](https://github.com/google-gemini/gemini-cli/pull/15714)
+- feat(hooks): simplify hook firing with HookSystem wrapper methods by @ved015
+  in [#15982](https://github.com/google-gemini/gemini-cli/pull/15982)
+- Add exp.gws_experiment field to LogEventEntry by @gsquared94 in
+  [#16062](https://github.com/google-gemini/gemini-cli/pull/16062)
+- Revert "feat(admin): implement extensions disabled" by @chrstnb in
+  [#16082](https://github.com/google-gemini/gemini-cli/pull/16082)
+- feat(core): Decouple enabling hooks UI from subsystem. by @joshualitt in
+  [#16074](https://github.com/google-gemini/gemini-cli/pull/16074)
+- docs: add docs for hooks + extensions by @abhipatel12 in
+  [#16073](https://github.com/google-gemini/gemini-cli/pull/16073)
+- feat(core): Preliminary changes for subagent model routing. by @joshualitt in
+  [#16035](https://github.com/google-gemini/gemini-cli/pull/16035)
+- Optimize CI workflow: Parallelize jobs and cache linters by @NTaylorMullen in
+  [#16054](https://github.com/google-gemini/gemini-cli/pull/16054)
+- Add option to fallback for capacity errors in ProQuotaDi… by @sehoon38 in
+  [#16050](https://github.com/google-gemini/gemini-cli/pull/16050)
+- feat: add confirmation details support + jsonrpc vs http rest support by
+  @adamfweidman in
+  [#16079](https://github.com/google-gemini/gemini-cli/pull/16079)
+- fix(workflows): fix and limit labels for pr-triage.sh script by @jacob314 in
+  [#16096](https://github.com/google-gemini/gemini-cli/pull/16096)
+- Fix and rename introspection agent -> cli help agent by @scidomino in
+  [#16097](https://github.com/google-gemini/gemini-cli/pull/16097)
+- Docs: Changelogs update 20260105 by @jkcinouye in
+  [#15937](https://github.com/google-gemini/gemini-cli/pull/15937)
+- enable cli_help agent by default by @scidomino in
+  [#16100](https://github.com/google-gemini/gemini-cli/pull/16100)
+- Optimize json-output tests with mock responses by @NTaylorMullen in
+  [#16102](https://github.com/google-gemini/gemini-cli/pull/16102)
+- Fix CI for forks by @scidomino in
+  [#16113](https://github.com/google-gemini/gemini-cli/pull/16113)
+- Reduce nags about PRs that reference issues but don't fix them. by @jacob314
+  in [#16112](https://github.com/google-gemini/gemini-cli/pull/16112)
+- feat(cli): add filepath autosuggestion after slash commands by @jasmeetsb in
+  [#14738](https://github.com/google-gemini/gemini-cli/pull/14738)
+- Add upgrade option for paid users by @cayden-google in
+  [#15978](https://github.com/google-gemini/gemini-cli/pull/15978)
+- [Skills] UX Polishing: Transparent feedback and CLI refinements by
+  @NTaylorMullen in
+  [#15954](https://github.com/google-gemini/gemini-cli/pull/15954)
+- Polish: Move 'Failed to load skills' warning to debug logs by @NTaylorMullen
+  in [#16142](https://github.com/google-gemini/gemini-cli/pull/16142)
+- feat(cli): export chat history in /bug and prefill GitHub issue by
+  @NTaylorMullen in
+  [#16115](https://github.com/google-gemini/gemini-cli/pull/16115)
+- bug(core): fix issue with overrides to bases. by @joshualitt in
+  [#15255](https://github.com/google-gemini/gemini-cli/pull/15255)
+- enableInteractiveShell for external tooling relying on a2a server by
+  @DavidAPierce in
+  [#16080](https://github.com/google-gemini/gemini-cli/pull/16080)
+- Reapply "feat(admin): implement extensions disabled" (#16082) by @skeshive in
+  [#16109](https://github.com/google-gemini/gemini-cli/pull/16109)
+- bug(core): Fix spewie getter in hookTranslator.ts by @joshualitt in
+  [#16108](https://github.com/google-gemini/gemini-cli/pull/16108)
+- feat(hooks): add mcp_context to BeforeTool and AfterTool hook inputs by @vrv
+  in [#15656](https://github.com/google-gemini/gemini-cli/pull/15656)
+- Add extension linking capabilities in cli by @kevinjwang1 in
+  [#16040](https://github.com/google-gemini/gemini-cli/pull/16040)
+- Update the page's title to be consistent and show in site. by @kschaab in
+  [#16174](https://github.com/google-gemini/gemini-cli/pull/16174)
+- docs: correct typo in bufferFastReturn JSDoc ("accomodate" → "accommodate") by
+  @minglu7 in [#16056](https://github.com/google-gemini/gemini-cli/pull/16056)
+- fix: typo in MCP servers settings description by @alphanota in
+  [#15929](https://github.com/google-gemini/gemini-cli/pull/15929)
+- fix: yolo should auto allow redirection by @abhipatel12 in
+  [#16183](https://github.com/google-gemini/gemini-cli/pull/16183)
+- fix(cli): disableYoloMode shouldn't enforce default approval mode against args
+  by @psinha40898 in
+  [#16155](https://github.com/google-gemini/gemini-cli/pull/16155)
+- feat: add native Sublime Text support to IDE detection by @phreakocious in
+  [#16083](https://github.com/google-gemini/gemini-cli/pull/16083)
+- refactor(core): extract ToolModificationHandler from scheduler by @abhipatel12
+  in [#16118](https://github.com/google-gemini/gemini-cli/pull/16118)
+- Add support for Antigravity terminal in terminal setup utility by @raky291 in
+  [#16051](https://github.com/google-gemini/gemini-cli/pull/16051)
+- feat(core): Wire up model routing to subagents. by @joshualitt in
+  [#16043](https://github.com/google-gemini/gemini-cli/pull/16043)
+- feat(cli): add /agents slash command to list available agents by @adamfweidman
+  in [#16182](https://github.com/google-gemini/gemini-cli/pull/16182)
+- docs(cli): fix includeDirectories nesting in configuration.md by @maru0804 in
+  [#15067](https://github.com/google-gemini/gemini-cli/pull/15067)
+- feat: implement file system reversion utilities for rewind by @Adib234 in
+  [#15715](https://github.com/google-gemini/gemini-cli/pull/15715)
+- Always enable redaction in GitHub actions. by @gundermanc in
+  [#16200](https://github.com/google-gemini/gemini-cli/pull/16200)
+- fix: remove unsupported 'enabled' key from workflow config by @Han5991 in
+  [#15611](https://github.com/google-gemini/gemini-cli/pull/15611)
+- docs: Remove redundant and duplicate documentation files by @liqzheng in
+  [#14699](https://github.com/google-gemini/gemini-cli/pull/14699)
+- docs: shorten run command and use published version by @dsherret in
+  [#16172](https://github.com/google-gemini/gemini-cli/pull/16172)
+- test(command-registry): increase initialization test timeout by @wszqkzqk in
+  [#15979](https://github.com/google-gemini/gemini-cli/pull/15979)
+- Ensure TERM is set to xterm-256color by @falouu in
+  [#15828](https://github.com/google-gemini/gemini-cli/pull/15828)
+- The telemetry.js script should handle paths that contain spaces by @JohnJAS in
+  [#12078](https://github.com/google-gemini/gemini-cli/pull/12078)
+- ci: guard links workflow from running on forks by @wtanaka in
+  [#15461](https://github.com/google-gemini/gemini-cli/pull/15461)
+- ci: guard nightly release workflow from running on forks by @wtanaka in
+  [#15463](https://github.com/google-gemini/gemini-cli/pull/15463)
+- Support @ suggestions for subagenets by @sehoon38 in
+  [#16201](https://github.com/google-gemini/gemini-cli/pull/16201)
+- feat(hooks): Support explicit stop and block execution control in model hooks
+  by @SandyTao520 in
+  [#15947](https://github.com/google-gemini/gemini-cli/pull/15947)
+- Refine Gemini 3 system instructions to reduce model verbosity by
+  @NTaylorMullen in
+  [#16139](https://github.com/google-gemini/gemini-cli/pull/16139)
+- chore: clean up unused models and use consts by @sehoon38 in
+  [#16246](https://github.com/google-gemini/gemini-cli/pull/16246)
+- Always enable bracketed paste by @scidomino in
+  [#16179](https://github.com/google-gemini/gemini-cli/pull/16179)
+- refactor: migrate clearCommand hook calls to HookSystem by @ved015 in
+  [#16157](https://github.com/google-gemini/gemini-cli/pull/16157)
+- refactor: migrate app containter hook calls to hook system by @ishaanxgupta in
+  [#16161](https://github.com/google-gemini/gemini-cli/pull/16161)
+- Show settings source in extensions lists by @chrstnb in
+  [#16207](https://github.com/google-gemini/gemini-cli/pull/16207)
+- feat(skills): add pr-creator skill and enable skills by @NTaylorMullen in
+  [#16232](https://github.com/google-gemini/gemini-cli/pull/16232)
+- fix: handle Shift+Space in Kitty keyboard protocol terminals by @tt-a1i in
+  [#15767](https://github.com/google-gemini/gemini-cli/pull/15767)
+- feat(core, ui): Add /agents refresh command. by @joshualitt in
+  [#16204](https://github.com/google-gemini/gemini-cli/pull/16204)
+- feat(core): add local experiments override via GEMINI_EXP by @kevin-ramdass in
+  [#16181](https://github.com/google-gemini/gemini-cli/pull/16181)
+- feat(ui): reduce home directory warning noise and add opt-out setting by
+  @NTaylorMullen in
+  [#16229](https://github.com/google-gemini/gemini-cli/pull/16229)
+- refactor: migrate chatCompressionService to use HookSystem by @ved015 in
+  [#16259](https://github.com/google-gemini/gemini-cli/pull/16259)
+- fix: properly use systemMessage for hooks in UI by @jackwotherspoon in
+  [#16250](https://github.com/google-gemini/gemini-cli/pull/16250)
+- Infer modifyOtherKeys support by @scidomino in
+  [#16270](https://github.com/google-gemini/gemini-cli/pull/16270)
+- feat(core): Cache ignore instances for performance by @EricRahm in
+  [#16185](https://github.com/google-gemini/gemini-cli/pull/16185)
+- feat: apply remote admin settings (no-op) by @skeshive in
+  [#16106](https://github.com/google-gemini/gemini-cli/pull/16106)
+- Autogenerate docs/cli/settings.md docs/getting-started/configuration.md was
+  already autogenerated but settings.md was not. by @jacob314 in
+  [#14408](https://github.com/google-gemini/gemini-cli/pull/14408)
+- refactor(config): remove legacy V1 settings migration logic by @galz10 in
+  [#16252](https://github.com/google-gemini/gemini-cli/pull/16252)
+- Fix an issue where the agent stops prematurely by @gundermanc in
+  [#16269](https://github.com/google-gemini/gemini-cli/pull/16269)
+- Update system prompt to prefer non-interactive commands by @NTaylorMullen in
+  [#16117](https://github.com/google-gemini/gemini-cli/pull/16117)
+- Update ink version to 6.4.7 by @jacob314 in
+  [#16284](https://github.com/google-gemini/gemini-cli/pull/16284)
+- Support for Built-in Agent Skills by @NTaylorMullen in
+  [#16045](https://github.com/google-gemini/gemini-cli/pull/16045)
+- fix(skills): remove "Restart required" message from non-interactive commands
+  by @NTaylorMullen in
+  [#16307](https://github.com/google-gemini/gemini-cli/pull/16307)
+- remove unused sessionHookTriggers and exports by @ved015 in
+  [#16324](https://github.com/google-gemini/gemini-cli/pull/16324)
+- Triage action cleanup by @bdmorgan in
+  [#16319](https://github.com/google-gemini/gemini-cli/pull/16319)
+- fix: Add event-driven trigger to issue triage workflow by @bdmorgan in
+  [#16334](https://github.com/google-gemini/gemini-cli/pull/16334)
+- fix(workflows): resolve triage workflow failures and actionlint errors by
+  @bdmorgan in [#16338](https://github.com/google-gemini/gemini-cli/pull/16338)
+- docs: add note about experimental hooks by @abhipatel12 in
+  [#16337](https://github.com/google-gemini/gemini-cli/pull/16337)
+- feat(cli): implement passive activity logger for session analysis by
+  @SandyTao520 in
+  [#15829](https://github.com/google-gemini/gemini-cli/pull/15829)
+- feat(cli): add /chat debug command for nightly builds by @abhipatel12 in
+  [#16339](https://github.com/google-gemini/gemini-cli/pull/16339)
+- style: format pr-creator skill by @NTaylorMullen in
+  [#16381](https://github.com/google-gemini/gemini-cli/pull/16381)
+- feat(cli): Hooks enable-all/disable-all feature with dynamic status by
+  @AbdulTawabJuly in
+  [#15552](https://github.com/google-gemini/gemini-cli/pull/15552)
+- fix(core): ensure silent local subagent delegation while allowing remote
+  confirmation by @adamfweidman in
+  [#16395](https://github.com/google-gemini/gemini-cli/pull/16395)
+- Markdown w/ Frontmatter Agent Parser by @sehoon38 in
+  [#16094](https://github.com/google-gemini/gemini-cli/pull/16094)
+- Fix crash on unicode character by @chrstnb in
+  [#16420](https://github.com/google-gemini/gemini-cli/pull/16420)
+- Attempt to resolve OOM w/ useMemo on history items by @chrstnb in
+  [#16424](https://github.com/google-gemini/gemini-cli/pull/16424)
+- fix(core): ensure sub-agent schema and prompt refresh during runtime by
+  @adamfweidman in
+  [#16409](https://github.com/google-gemini/gemini-cli/pull/16409)
+- Update extension examples by @chrstnb in
+  [#16274](https://github.com/google-gemini/gemini-cli/pull/16274)
+- revert the change that was recently added from a fix by @sehoon38 in
+  [#16390](https://github.com/google-gemini/gemini-cli/pull/16390)
+- Add other hook wrapper methods to hooksystem by @ved015 in
+  [#16361](https://github.com/google-gemini/gemini-cli/pull/16361)
+- feat: introduce useRewindLogic hook for conversation history navigation by
+  @Adib234 in [#15716](https://github.com/google-gemini/gemini-cli/pull/15716)
+- docs: Fix formatting issue in memport documentation by @wanglc02 in
+  [#14774](https://github.com/google-gemini/gemini-cli/pull/14774)
+- fix(policy): enhance shell command safety and parsing by @allenhutchison in
+  [#15034](https://github.com/google-gemini/gemini-cli/pull/15034)
+- fix(core): avoid 'activate_skill' re-registration warning by @NTaylorMullen in
+  [#16398](https://github.com/google-gemini/gemini-cli/pull/16398)
+- perf(workflows): optimize PR triage script for faster execution by @bdmorgan
+  in [#16355](https://github.com/google-gemini/gemini-cli/pull/16355)
+- feat(admin): prompt user to restart the CLI if they change auth to oauth
+  mid-session or don't have auth type selected at start of session by @skeshive
+  in [#16426](https://github.com/google-gemini/gemini-cli/pull/16426)
+- Update cli-help agent's system prompt in sub-agents section by @sehoon38 in
+  [#16441](https://github.com/google-gemini/gemini-cli/pull/16441)
+- Revert "Update extension examples" by @chrstnb in
+  [#16442](https://github.com/google-gemini/gemini-cli/pull/16442)
+- Fix: add back fastreturn support by @scidomino in
+  [#16440](https://github.com/google-gemini/gemini-cli/pull/16440)
+- feat(a2a): Introduce /memory command for a2a server by @cocosheng-g in
+  [#14456](https://github.com/google-gemini/gemini-cli/pull/14456)
+- docs: fix broken internal link by using relative path by @Gong-Mi in
+  [#15371](https://github.com/google-gemini/gemini-cli/pull/15371)
+- migrate yolo/auto-edit keybindings by @scidomino in
+  [#16457](https://github.com/google-gemini/gemini-cli/pull/16457)
+- feat(cli): add install and uninstall commands for skills by @NTaylorMullen in
+  [#16377](https://github.com/google-gemini/gemini-cli/pull/16377)
+- feat(ui): use Tab to switch focus between shell and input by @jacob314 in
+  [#14332](https://github.com/google-gemini/gemini-cli/pull/14332)
+- feat(core): support shipping built-in skills with the CLI by @NTaylorMullen in
+  [#16300](https://github.com/google-gemini/gemini-cli/pull/16300)
+- Collect hardware details telemetry. by @gundermanc in
+  [#16119](https://github.com/google-gemini/gemini-cli/pull/16119)
+- feat(agents): improve UI feedback and parser reliability by @NTaylorMullen in
+  [#16459](https://github.com/google-gemini/gemini-cli/pull/16459)
+- Migrate keybindings by @scidomino in
+  [#16460](https://github.com/google-gemini/gemini-cli/pull/16460)
+- feat(cli): cleanup activity logs alongside session files by @SandyTao520 in
+  [#16399](https://github.com/google-gemini/gemini-cli/pull/16399)
+- feat(cli): implement dynamic terminal tab titles for CLI status by
+  @NTaylorMullen in
+  [#16378](https://github.com/google-gemini/gemini-cli/pull/16378)
+- feat(core): add disableLLMCorrection setting to skip auto-correction in edit
+  tools by @SandyTao520 in
+  [#16000](https://github.com/google-gemini/gemini-cli/pull/16000)
+- fix: Set both tab and window title instead of just window title by
+  @NTaylorMullen in
+  [#16464](https://github.com/google-gemini/gemini-cli/pull/16464)
+- fix(policy): ensure MCP policies match unqualified names in non-interactive
+  mode by @NTaylorMullen in
+  [#16490](https://github.com/google-gemini/gemini-cli/pull/16490)
+- fix(cli): refine 'Action Required' indicator and focus hints by @NTaylorMullen
+  in [#16497](https://github.com/google-gemini/gemini-cli/pull/16497)
+- Refactor beforeAgent and afterAgent hookEvents to follow desired output by
+  @ved015 in [#16495](https://github.com/google-gemini/gemini-cli/pull/16495)
+- feat(agents): clarify mandatory YAML frontmatter for sub-agents by
+  @NTaylorMullen in
+  [#16515](https://github.com/google-gemini/gemini-cli/pull/16515)
+- docs(telemetry): add Google Cloud Monitoring dashboard documentation by @jerop
+  in [#16520](https://github.com/google-gemini/gemini-cli/pull/16520)
+- Implement support for subagents as extensions. by @gundermanc in
+  [#16473](https://github.com/google-gemini/gemini-cli/pull/16473)
+- refactor: make baseTimestamp optional in addItem and remove redundant calls by
+  @sehoon38 in [#16471](https://github.com/google-gemini/gemini-cli/pull/16471)
+- Improve key binding names and descriptions by @scidomino in
+  [#16529](https://github.com/google-gemini/gemini-cli/pull/16529)
+- feat(core, cli): Add support for agents in settings.json. by @joshualitt in
+  [#16433](https://github.com/google-gemini/gemini-cli/pull/16433)
+- fix(cli): fix 'gemini skills install' unknown argument error by @NTaylorMullen
+  in [#16537](https://github.com/google-gemini/gemini-cli/pull/16537)
+- chore(ui): optimize AgentsStatus layout with dense list style and group
+  separation by @adamfweidman in
+  [#16545](https://github.com/google-gemini/gemini-cli/pull/16545)
+- fix(cli): allow @ file selector on slash command lines by @galz10 in
+  [#16370](https://github.com/google-gemini/gemini-cli/pull/16370)
+- fix(ui): resolve sticky header regression in tool messages by @jacob314 in
+  [#16514](https://github.com/google-gemini/gemini-cli/pull/16514)
+- feat(core): Align internal agent settings with configs exposed through
+  settings.json by @joshualitt in
+  [#16458](https://github.com/google-gemini/gemini-cli/pull/16458)
+- fix(cli): copy uses OSC52 only in SSH/WSL by @assagman in
+  [#16554](https://github.com/google-gemini/gemini-cli/pull/16554)
+- docs(skills): clarify skill directory structure and file location by
+  @NTaylorMullen in
+  [#16532](https://github.com/google-gemini/gemini-cli/pull/16532)
+- Fix: make ctrl+x use preferred editor by @scidomino in
+  [#16556](https://github.com/google-gemini/gemini-cli/pull/16556)
+- fix(core): Resolve race condition in tool response reporting by @abhipatel12
+  in [#16557](https://github.com/google-gemini/gemini-cli/pull/16557)
+- feat(ui): highlight persist mode status in ModelDialog by @sehoon38 in
+  [#16483](https://github.com/google-gemini/gemini-cli/pull/16483)
+- refactor: clean up A2A task output for users and LLMs by @adamfweidman in
+  [#16561](https://github.com/google-gemini/gemini-cli/pull/16561)
+- feat(core/ui): enhance retry mechanism and UX by @sehoon38 in
+  [#16489](https://github.com/google-gemini/gemini-cli/pull/16489)
+- Modernize MaxSizedBox to use and ResizeObservers by @jacob314 in
+  [#16565](https://github.com/google-gemini/gemini-cli/pull/16565)
+- Behavioral evals framework. by @gundermanc in
+  [#16047](https://github.com/google-gemini/gemini-cli/pull/16047)
+- Aggregate test results. by @gundermanc in
+  [#16581](https://github.com/google-gemini/gemini-cli/pull/16581)
+- feat(admin): support admin-enforced settings for Agent Skills by
+  @NTaylorMullen in
+  [#16406](https://github.com/google-gemini/gemini-cli/pull/16406)
+- fix(patch): cherry-pick cfdc4cf to release/v0.25.0-preview.0-pr-16759 to patch
+  version v0.25.0-preview.0 and create version 0.25.0-preview.1 by
+  @gemini-cli-robot in
+  [#16866](https://github.com/google-gemini/gemini-cli/pull/16866)
+- Patch #16730 into v0.25.0 preview by @chrstnb in
+  [#16882](https://github.com/google-gemini/gemini-cli/pull/16882)
+- fix(patch): cherry-pick 3b55581 to release/v0.25.0-preview.2-pr-16506 to patch
+  version v0.25.0-preview.2 and create version 0.25.0-preview.3 by
+  @gemini-cli-robot in
+  [#17098](https://github.com/google-gemini/gemini-cli/pull/17098)
 
-**Full Changelog**:
-https://github.com/google-gemini/gemini-cli/compare/v0.21.3...v0.22.0
+**Full changelog**:
+https://github.com/google-gemini/gemini-cli/compare/v0.24.5...v0.25.0

docs/changelogs/preview.md

@@ -1,6 +1,6 @@
-# Preview release: Release v0.23.0-preview.0
+# Preview release: Release v0.26.0-preview.0
 
-Released: December 22, 2025
+Released: January 21, 2026
 
 Our preview release includes the latest, new, and experimental features. This
 release may not be as stable as our [latest weekly release](latest.md).
@@ -11,121 +11,322 @@ To install the preview release:
 npm install -g @google/gemini-cli@preview
 ```
 
+## Highlights
+
+- **Skills and Agents:** Improvements to the `activate_skill` tool and skill
+  management. Experimental Agent Skills support.
+- **UI/UX:** Addition of a Rewind Confirmation dialog and Viewer component.
+- **Extensions:** Experimental setting for extension configuration.
+- **Bug Fixes and Stability:** PDF token estimation fix and improvements to
+  scheduled issue triage.
+
 ## What's Changed
 
-- Code assist service metrics. by @gundermanc in
-  https://github.com/google-gemini/gemini-cli/pull/15024
-- chore/release: bump version to 0.21.0-nightly.20251216.bb0c0d8ee by
-  @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/15121
-- Docs by @Roaimkhan in https://github.com/google-gemini/gemini-cli/pull/15103
-- Use official ACP SDK and support HTTP/SSE based MCP servers by @SteffenDE in
-  https://github.com/google-gemini/gemini-cli/pull/13856
-- Remove foreground for themes other than shades of purple and holiday. by
-  @jacob314 in https://github.com/google-gemini/gemini-cli/pull/14606
-- chore: remove repo specific tips by @jackwotherspoon in
-  https://github.com/google-gemini/gemini-cli/pull/15164
-- chore: remove user query from footer in debug mode by @jackwotherspoon in
-  https://github.com/google-gemini/gemini-cli/pull/15169
-- Disallow unnecessary awaits. by @gundermanc in
-  https://github.com/google-gemini/gemini-cli/pull/15172
-- Add one to the padding in settings dialog to avoid flicker. by @jacob314 in
-  https://github.com/google-gemini/gemini-cli/pull/15173
-- feat(core): introduce remote agent infrastructure and rename local executor by
-  @adamfweidman in https://github.com/google-gemini/gemini-cli/pull/15110
-- feat(cli): Add `/auth logout` command to clear credentials and auth state by
-  @CN-Scars in https://github.com/google-gemini/gemini-cli/pull/13383
-- (fix) Automated pr labeler by @DaanVersavel in
-  https://github.com/google-gemini/gemini-cli/pull/14885
-- feat: launch Gemini 3 Flash in Gemini CLI ⚡️⚡️⚡️ by @scidomino in
-  https://github.com/google-gemini/gemini-cli/pull/15196
-- Refactor: Migrate console.error in ripGrep.ts to debugLogger by @Adib234 in
-  https://github.com/google-gemini/gemini-cli/pull/15201
-- chore: update a2a-js to 0.3.7 by @adamfweidman in
-  https://github.com/google-gemini/gemini-cli/pull/15197
-- chore(core): remove redundant isModelAvailabilityServiceEnabled toggle and
-  clean up dead code by @adamfweidman in
-  https://github.com/google-gemini/gemini-cli/pull/15207
-- feat(core): Late resolve `GenerateContentConfig`s and reduce mutation. by
-  @joshualitt in https://github.com/google-gemini/gemini-cli/pull/14920
-- Respect previewFeatures value from the remote flag if undefined by @sehoon38
-  in https://github.com/google-gemini/gemini-cli/pull/15214
-- feat(ui): add Windows clipboard image support and Alt+V paste workaround by
-  @jacob314 in https://github.com/google-gemini/gemini-cli/pull/15218
-- chore(core): remove legacy fallback flags and migrate loop detection by
-  @adamfweidman in https://github.com/google-gemini/gemini-cli/pull/15213
-- fix(ui): Prevent eager slash command completion hiding sibling commands by
-  @SandyTao520 in https://github.com/google-gemini/gemini-cli/pull/15224
-- Docs: Update Changelog for Dec 17, 2025 by @jkcinouye in
-  https://github.com/google-gemini/gemini-cli/pull/15204
-- Code Assist backend telemetry for user accept/reject of suggestions by
-  @gundermanc in https://github.com/google-gemini/gemini-cli/pull/15206
-- fix(cli): correct initial history length handling for chat commands by
-  @SandyTao520 in https://github.com/google-gemini/gemini-cli/pull/15223
-- chore/release: bump version to 0.21.0-nightly.20251218.739c02bd6 by
-  @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/15231
-- Change detailed model stats to use a new shared Table class to resolve
-  robustness issues. by @jacob314 in
-  https://github.com/google-gemini/gemini-cli/pull/15208
-- feat: add agent toml parser by @abhipatel12 in
-  https://github.com/google-gemini/gemini-cli/pull/15112
-- Add core tool that adds all context from the core package. by @jacob314 in
-  https://github.com/google-gemini/gemini-cli/pull/15238
-- (docs): Add reference section to hooks documentation by @abhipatel12 in
-  https://github.com/google-gemini/gemini-cli/pull/15159
-- feat(hooks): add support for friendly names and descriptions by @abhipatel12
-  in https://github.com/google-gemini/gemini-cli/pull/15174
-- feat: Detect background color by @jacob314 in
-  https://github.com/google-gemini/gemini-cli/pull/15132
-- add 3.0 to allowed sensitive keywords by @scidomino in
-  https://github.com/google-gemini/gemini-cli/pull/15276
-- feat: Pass additional environment variables to shell execution by @galz10 in
-  https://github.com/google-gemini/gemini-cli/pull/15160
-- Remove unused code by @scidomino in
-  https://github.com/google-gemini/gemini-cli/pull/15290
-- Handle all 429 as retryableQuotaError by @sehoon38 in
-  https://github.com/google-gemini/gemini-cli/pull/15288
-- Remove unnecessary dependencies by @scidomino in
-  https://github.com/google-gemini/gemini-cli/pull/15291
-- fix: prevent infinite loop in prompt completion on error by @galz10 in
-  https://github.com/google-gemini/gemini-cli/pull/14548
-- fix(ui): show command suggestions even on perfect match and sort them by
-  @SandyTao520 in https://github.com/google-gemini/gemini-cli/pull/15287
-- feat(hooks): reduce log verbosity and improve error reporting in UI by
-  @abhipatel12 in https://github.com/google-gemini/gemini-cli/pull/15297
-- feat: simplify tool confirmation labels for better UX by @NTaylorMullen in
-  https://github.com/google-gemini/gemini-cli/pull/15296
-- chore/release: bump version to 0.21.0-nightly.20251219.70696e364 by
-  @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/15301
-- feat(core): Implement JIT context memory loading and UI sync by @SandyTao520
-  in https://github.com/google-gemini/gemini-cli/pull/14469
-- feat(ui): Put "Allow for all future sessions" behind a setting off by default.
-  by @jacob314 in https://github.com/google-gemini/gemini-cli/pull/15322
-- fix(cli):change the placeholder of input during the shell mode by
-  @JayadityaGit in https://github.com/google-gemini/gemini-cli/pull/15135
-- Validate OAuth resource parameter matches MCP server URL by @galz10 in
-  https://github.com/google-gemini/gemini-cli/pull/15289
-- docs(cli): add System Prompt Override (GEMINI_SYSTEM_MD) by @ashmod in
-  https://github.com/google-gemini/gemini-cli/pull/9515
-- more robust command parsing logs by @scidomino in
-  https://github.com/google-gemini/gemini-cli/pull/15339
-- Introspection agent demo by @scidomino in
-  https://github.com/google-gemini/gemini-cli/pull/15232
-- fix(core): sanitize hook command expansion and prevent injection by
-  @SandyTao520 in https://github.com/google-gemini/gemini-cli/pull/15343
-- fix(folder trust): add validation for trusted folder level by @adamfweidman in
-  https://github.com/google-gemini/gemini-cli/pull/12215
-- fix(cli): fix right border overflow in trust dialogs by @galz10 in
-  https://github.com/google-gemini/gemini-cli/pull/15350
-- fix(policy): fix bug where accepting-edits continued after it was turned off
-  by @jacob314 in https://github.com/google-gemini/gemini-cli/pull/15351
-- fix: prevent infinite relaunch loop when --resume fails (#14941) by @Ying-xi
-  in https://github.com/google-gemini/gemini-cli/pull/14951
-- chore/release: bump version to 0.21.0-nightly.20251220.41a1a3eed by
-  @gemini-cli-robot in https://github.com/google-gemini/gemini-cli/pull/15352
-- feat(telemetry): add clearcut logging for hooks by @abhipatel12 in
-  https://github.com/google-gemini/gemini-cli/pull/15405
-- fix(core): Add `.geminiignore` support to SearchText tool by @xyrolle in
-  https://github.com/google-gemini/gemini-cli/pull/13763
+- fix: PDF token estimation
+  ([#16494](https://github.com/google-gemini/gemini-cli/pull/16494)) by
+  @korade-krushna in
+  [#16527](https://github.com/google-gemini/gemini-cli/pull/16527)
+- chore(release): bump version to 0.26.0-nightly.20260114.bb6c57414 by
+  @gemini-cli-robot in
+  [#16604](https://github.com/google-gemini/gemini-cli/pull/16604)
+- docs: clarify F12 to open debug console by @jackwotherspoon in
+  [#16570](https://github.com/google-gemini/gemini-cli/pull/16570)
+- docs: Remove .md extension from internal links in architecture.md by
+  @medic-code in
+  [#12899](https://github.com/google-gemini/gemini-cli/pull/12899)
+- Add an experimental setting for extension config by @chrstnb in
+  [#16506](https://github.com/google-gemini/gemini-cli/pull/16506)
+- feat: add Rewind Confirmation dialog and Rewind Viewer component by @Adib234
+  in [#15717](https://github.com/google-gemini/gemini-cli/pull/15717)
+- fix(a2a): Don't throw errors for GeminiEventType Retry and InvalidStream. by
+  @ehedlund in [#16541](https://github.com/google-gemini/gemini-cli/pull/16541)
+- prefactor: add rootCommands as array so it can be used for policy parsing by
+  @abhipatel12 in
+  [#16640](https://github.com/google-gemini/gemini-cli/pull/16640)
+- remove unnecessary \x7f key bindings by @scidomino in
+  [#16646](https://github.com/google-gemini/gemini-cli/pull/16646)
+- docs(skills): use body-file in pr-creator skill for better reliability by
+  @abhipatel12 in
+  [#16642](https://github.com/google-gemini/gemini-cli/pull/16642)
+- chore(automation): recursive labeling for workstream descendants by @bdmorgan
+  in [#16609](https://github.com/google-gemini/gemini-cli/pull/16609)
+- feat: introduce 'skill-creator' built-in skill and CJS management tools by
+  @NTaylorMullen in
+  [#16394](https://github.com/google-gemini/gemini-cli/pull/16394)
+- chore(automation): remove automated PR size and complexity labeler by
+  @bdmorgan in [#16648](https://github.com/google-gemini/gemini-cli/pull/16648)
+- refactor(skills): replace 'project' with 'workspace' scope by @NTaylorMullen
+  in [#16380](https://github.com/google-gemini/gemini-cli/pull/16380)
+- Docs: Update release notes for 1/13/2026 by @jkcinouye in
+  [#16583](https://github.com/google-gemini/gemini-cli/pull/16583)
+- Simplify paste handling by @scidomino in
+  [#16654](https://github.com/google-gemini/gemini-cli/pull/16654)
+- chore(automation): improve scheduled issue triage discovery and throughput by
+  @bdmorgan in [#16652](https://github.com/google-gemini/gemini-cli/pull/16652)
+- fix(acp): run exit cleanup when stdin closes by @codefromthecrypt in
+  [#14953](https://github.com/google-gemini/gemini-cli/pull/14953)
+- feat(scheduler): add types needed for event driven scheduler by @abhipatel12
+  in [#16641](https://github.com/google-gemini/gemini-cli/pull/16641)
+- Remove unused rewind key binding by @scidomino in
+  [#16659](https://github.com/google-gemini/gemini-cli/pull/16659)
+- Remove sequence binding by @scidomino in
+  [#16664](https://github.com/google-gemini/gemini-cli/pull/16664)
+- feat(cli): undeprecate the --prompt flag by @alexaustin007 in
+  [#13981](https://github.com/google-gemini/gemini-cli/pull/13981)
+- chore: update dependabot configuration by @cosmopax in
+  [#13507](https://github.com/google-gemini/gemini-cli/pull/13507)
+- feat(config): add 'auto' alias for default model selection by @sehoon38 in
+  [#16661](https://github.com/google-gemini/gemini-cli/pull/16661)
+- Enable & disable agents by @sehoon38 in
+  [#16225](https://github.com/google-gemini/gemini-cli/pull/16225)
+- cleanup: Improve keybindings by @scidomino in
+  [#16672](https://github.com/google-gemini/gemini-cli/pull/16672)
+- Add timeout for shell-utils to prevent hangs. by @jacob314 in
+  [#16667](https://github.com/google-gemini/gemini-cli/pull/16667)
+- feat(plan): add experimental plan flag by @jerop in
+  [#16650](https://github.com/google-gemini/gemini-cli/pull/16650)
+- feat(cli): add security consent prompts for skill installation by
+  @NTaylorMullen in
+  [#16549](https://github.com/google-gemini/gemini-cli/pull/16549)
+- fix: replace 3 consecutive periods with ellipsis character by @Vist233 in
+  [#16587](https://github.com/google-gemini/gemini-cli/pull/16587)
+- chore(automation): ensure status/need-triage is applied and never cleared
+  automatically by @bdmorgan in
+  [#16657](https://github.com/google-gemini/gemini-cli/pull/16657)
+- fix: Handle colons in skill description frontmatter by @maru0804 in
+  [#16345](https://github.com/google-gemini/gemini-cli/pull/16345)
+- refactor(core): harden skill frontmatter parsing by @NTaylorMullen in
+  [#16705](https://github.com/google-gemini/gemini-cli/pull/16705)
+- feat(skills): add conflict detection and warnings for skill overrides by
+  @NTaylorMullen in
+  [#16709](https://github.com/google-gemini/gemini-cli/pull/16709)
+- feat(scheduler): add SchedulerStateManager for reactive tool state by
+  @abhipatel12 in
+  [#16651](https://github.com/google-gemini/gemini-cli/pull/16651)
+- chore(automation): enforce 'help wanted' label permissions and update
+  guidelines by @bdmorgan in
+  [#16707](https://github.com/google-gemini/gemini-cli/pull/16707)
+- fix(core): resolve circular dependency via tsconfig paths by @sehoon38 in
+  [#16730](https://github.com/google-gemini/gemini-cli/pull/16730)
+- chore/release: bump version to 0.26.0-nightly.20260115.6cb3ae4e0 by
+  @gemini-cli-robot in
+  [#16738](https://github.com/google-gemini/gemini-cli/pull/16738)
+- fix(automation): correct status/need-issue label matching wildcard by
+  @bdmorgan in [#16727](https://github.com/google-gemini/gemini-cli/pull/16727)
+- fix(automation): prevent label-enforcer loop by ignoring all bots by @bdmorgan
+  in [#16746](https://github.com/google-gemini/gemini-cli/pull/16746)
+- Add links to supported locations and minor fixes by @g-samroberts in
+  [#16476](https://github.com/google-gemini/gemini-cli/pull/16476)
+- feat(policy): add source tracking to policy rules by @allenhutchison in
+  [#16670](https://github.com/google-gemini/gemini-cli/pull/16670)
+- feat(automation): enforce '🔒 maintainer only' and fix bot loop by @bdmorgan
+  in [#16751](https://github.com/google-gemini/gemini-cli/pull/16751)
+- Make merged settings non-nullable and fix all lints related to that. by
+  @jacob314 in [#16647](https://github.com/google-gemini/gemini-cli/pull/16647)
+- fix(core): prevent ModelInfo event emission on aborted signal by @sehoon38 in
+  [#16752](https://github.com/google-gemini/gemini-cli/pull/16752)
+- Replace relative paths to fix website build by @chrstnb in
+  [#16755](https://github.com/google-gemini/gemini-cli/pull/16755)
+- Restricting to localhost by @cocosheng-g in
+  [#16548](https://github.com/google-gemini/gemini-cli/pull/16548)
+- fix(cli): add explicit dependency on color-convert by @sehoon38 in
+  [#16757](https://github.com/google-gemini/gemini-cli/pull/16757)
+- fix(automation): robust label enforcement with permission checks by @bdmorgan
+  in [#16762](https://github.com/google-gemini/gemini-cli/pull/16762)
+- fix(cli): prevent OOM crash by limiting file search traversal and adding
+  timeout by @galz10 in
+  [#16696](https://github.com/google-gemini/gemini-cli/pull/16696)
+- fix(cli): safely handle /dev/tty access on macOS by @korade-krushna in
+  [#16531](https://github.com/google-gemini/gemini-cli/pull/16531)
+- docs: clarify workspace test execution in GEMINI.md by @mattKorwel in
+  [#16764](https://github.com/google-gemini/gemini-cli/pull/16764)
+- Add support for running available commands prior to MCP servers loading by
+  @Adib234 in [#15596](https://github.com/google-gemini/gemini-cli/pull/15596)
+- feat(plan): add experimental 'plan' approval mode by @jerop in
+  [#16753](https://github.com/google-gemini/gemini-cli/pull/16753)
+- feat(scheduler): add functional awaitConfirmation utility by @abhipatel12 in
+  [#16721](https://github.com/google-gemini/gemini-cli/pull/16721)
+- fix(infra): update maintainer rollup label to 'workstream-rollup' by @bdmorgan
+  in [#16809](https://github.com/google-gemini/gemini-cli/pull/16809)
+- fix(infra): use GraphQL to detect direct parents in rollup workflow by
+  @bdmorgan in [#16811](https://github.com/google-gemini/gemini-cli/pull/16811)
+- chore(workflows): rename label-workstream-rollup workflow by @bdmorgan in
+  [#16818](https://github.com/google-gemini/gemini-cli/pull/16818)
+- skip simple-mcp-server.test.ts by @scidomino in
+  [#16842](https://github.com/google-gemini/gemini-cli/pull/16842)
+- Steer outer agent to use expert subagents when present by @gundermanc in
+  [#16763](https://github.com/google-gemini/gemini-cli/pull/16763)
+- Fix race condition by awaiting scheduleToolCalls by @chrstnb in
+  [#16759](https://github.com/google-gemini/gemini-cli/pull/16759)
+- cleanup: Organize key bindings by @scidomino in
+  [#16798](https://github.com/google-gemini/gemini-cli/pull/16798)
+- feat(core): Add generalist agent. by @joshualitt in
+  [#16638](https://github.com/google-gemini/gemini-cli/pull/16638)
+- perf(ui): optimize text buffer and highlighting for large inputs by
+  @NTaylorMullen in
+  [#16782](https://github.com/google-gemini/gemini-cli/pull/16782)
+- fix(core): fix PTY descriptor shell leak by @galz10 in
+  [#16773](https://github.com/google-gemini/gemini-cli/pull/16773)
+- feat(plan): enforce strict read-only policy and halt execution on violation by
+  @jerop in [#16849](https://github.com/google-gemini/gemini-cli/pull/16849)
+- remove need-triage label from bug_report template by @sehoon38 in
+  [#16864](https://github.com/google-gemini/gemini-cli/pull/16864)
+- fix(core): truncate large telemetry log entries by @sehoon38 in
+  [#16769](https://github.com/google-gemini/gemini-cli/pull/16769)
+- docs(extensions): add Agent Skills support and mark feature as experimental by
+  @NTaylorMullen in
+  [#16859](https://github.com/google-gemini/gemini-cli/pull/16859)
+- fix(core): surface warnings for invalid hook event names in configuration
+  ([#16788](https://github.com/google-gemini/gemini-cli/pull/16788)) by
+  @sehoon38 in [#16873](https://github.com/google-gemini/gemini-cli/pull/16873)
+- feat(plan): remove read_many_files from approval mode policies by @jerop in
+  [#16876](https://github.com/google-gemini/gemini-cli/pull/16876)
+- feat(admin): implement admin controls polling and restart prompt by @skeshive
+  in [#16627](https://github.com/google-gemini/gemini-cli/pull/16627)
+- Remove LRUCache class migrating to mnemoist by @jacob314 in
+  [#16872](https://github.com/google-gemini/gemini-cli/pull/16872)
+- feat(settings): rename negative settings to positive naming (disable* ->
+  enable*) by @afarber in
+  [#14142](https://github.com/google-gemini/gemini-cli/pull/14142)
+- refactor(cli): unify shell confirmation dialogs by @NTaylorMullen in
+  [#16828](https://github.com/google-gemini/gemini-cli/pull/16828)
+- feat(agent): enable agent skills by default by @NTaylorMullen in
+  [#16736](https://github.com/google-gemini/gemini-cli/pull/16736)
+- refactor(core): foundational truncation refactoring and token estimation
+  optimization by @NTaylorMullen in
+  [#16824](https://github.com/google-gemini/gemini-cli/pull/16824)
+- fix(hooks): enable /hooks disable to reliably stop single hooks by
+  @abhipatel12 in
+  [#16804](https://github.com/google-gemini/gemini-cli/pull/16804)
+- Don't commit unless user asks us to. by @gundermanc in
+  [#16902](https://github.com/google-gemini/gemini-cli/pull/16902)
+- chore: remove a2a-adapter and bump @a2a-js/sdk to 0.3.8 by @adamfweidman in
+  [#16800](https://github.com/google-gemini/gemini-cli/pull/16800)
+- fix: Show experiment values in settings UI for compressionThreshold by
+  @ishaanxgupta in
+  [#16267](https://github.com/google-gemini/gemini-cli/pull/16267)
+- feat(cli): replace relative keyboard shortcuts link with web URL by
+  @imaliabbas in
+  [#16479](https://github.com/google-gemini/gemini-cli/pull/16479)
+- fix(core): resolve PKCE length issue and stabilize OAuth redirect port by
+  @sehoon38 in [#16815](https://github.com/google-gemini/gemini-cli/pull/16815)
+- Delete rewind documentation for now by @Adib234 in
+  [#16932](https://github.com/google-gemini/gemini-cli/pull/16932)
+- Stabilize skill-creator CI and package format by @NTaylorMullen in
+  [#17001](https://github.com/google-gemini/gemini-cli/pull/17001)
+- Stabilize the git evals by @gundermanc in
+  [#16989](https://github.com/google-gemini/gemini-cli/pull/16989)
+- fix(core): attempt compression before context overflow check by @NTaylorMullen
+  in [#16914](https://github.com/google-gemini/gemini-cli/pull/16914)
+- Fix inverted logic. by @gundermanc in
+  [#17007](https://github.com/google-gemini/gemini-cli/pull/17007)
+- chore(scripts): add duplicate issue closer script and fix lint errors by
+  @bdmorgan in [#16997](https://github.com/google-gemini/gemini-cli/pull/16997)
+- docs: update README and config guide to reference Gemini 3 by @JayadityaGit in
+  [#15806](https://github.com/google-gemini/gemini-cli/pull/15806)
+- fix(cli): correct Homebrew installation detection by @kij in
+  [#14727](https://github.com/google-gemini/gemini-cli/pull/14727)
+- Demote git evals to nightly run. by @gundermanc in
+  [#17030](https://github.com/google-gemini/gemini-cli/pull/17030)
+- fix(cli): use OSC-52 clipboard copy in Windows Terminal by @Thomas-Shephard in
+  [#16920](https://github.com/google-gemini/gemini-cli/pull/16920)
+- Fix: Process all parts in response chunks when thought is first by @pyrytakala
+  in [#13539](https://github.com/google-gemini/gemini-cli/pull/13539)
+- fix(automation): fix jq quoting error in pr-triage.sh by @Kimsoo0119 in
+  [#16958](https://github.com/google-gemini/gemini-cli/pull/16958)
+- refactor(core): decouple scheduler into orchestration, policy, and
+  confirmation by @abhipatel12 in
+  [#16895](https://github.com/google-gemini/gemini-cli/pull/16895)
+- feat: add /introspect slash command by @NTaylorMullen in
+  [#17048](https://github.com/google-gemini/gemini-cli/pull/17048)
+- refactor(cli): centralize tool mapping and decouple legacy scheduler by
+  @abhipatel12 in
+  [#17044](https://github.com/google-gemini/gemini-cli/pull/17044)
+- fix(ui): ensure rationale renders before tool calls by @NTaylorMullen in
+  [#17043](https://github.com/google-gemini/gemini-cli/pull/17043)
+- fix(workflows): use author_association for maintainer check by @bdmorgan in
+  [#17060](https://github.com/google-gemini/gemini-cli/pull/17060)
+- fix return type of fireSessionStartEvent to defaultHookOutput by @ved015 in
+  [#16833](https://github.com/google-gemini/gemini-cli/pull/16833)
+- feat(cli): add experiment gate for event-driven scheduler by @abhipatel12 in
+  [#17055](https://github.com/google-gemini/gemini-cli/pull/17055)
+- feat(core): improve shell redirection transparency and security by
+  @NTaylorMullen in
+  [#16486](https://github.com/google-gemini/gemini-cli/pull/16486)
+- fix(core): deduplicate ModelInfo emission in GeminiClient by @NTaylorMullen in
+  [#17075](https://github.com/google-gemini/gemini-cli/pull/17075)
+- docs(themes): remove unsupported DiffModified color key by @jw409 in
+  [#17073](https://github.com/google-gemini/gemini-cli/pull/17073)
+- fix: update currentSequenceModel when modelChanged by @adamfweidman in
+  [#17051](https://github.com/google-gemini/gemini-cli/pull/17051)
+- feat(core): enhanced anchored iterative context compression with
+  self-verification by @rmedranollamas in
+  [#15710](https://github.com/google-gemini/gemini-cli/pull/15710)
+- Fix mcp instructions by @chrstnb in
+  [#16439](https://github.com/google-gemini/gemini-cli/pull/16439)
+- [A2A] Disable checkpointing if git is not installed by @cocosheng-g in
+  [#16896](https://github.com/google-gemini/gemini-cli/pull/16896)
+- feat(admin): set admin.skills.enabled based on advancedFeaturesEnabled setting
+  by @skeshive in
+  [#17095](https://github.com/google-gemini/gemini-cli/pull/17095)
+- Test coverage for hook exit code cases by @gundermanc in
+  [#17041](https://github.com/google-gemini/gemini-cli/pull/17041)
+- Revert "Revert "Update extension examples"" by @chrstnb in
+  [#16445](https://github.com/google-gemini/gemini-cli/pull/16445)
+- fix(core): Provide compact, actionable errors for agent delegation failures by
+  @SandyTao520 in
+  [#16493](https://github.com/google-gemini/gemini-cli/pull/16493)
+- fix: migrate BeforeModel and AfterModel hooks to HookSystem by @ved015 in
+  [#16599](https://github.com/google-gemini/gemini-cli/pull/16599)
+- feat(admin): apply admin settings to gemini skills/mcp/extensions commands by
+  @skeshive in [#17102](https://github.com/google-gemini/gemini-cli/pull/17102)
+- fix(core): update telemetry token count after session resume by @psinha40898
+  in [#15491](https://github.com/google-gemini/gemini-cli/pull/15491)
+- Demote the subagent test to nightly by @gundermanc in
+  [#17105](https://github.com/google-gemini/gemini-cli/pull/17105)
+- feat(plan): telemetry to track adoption and usage of plan mode by @Adib234 in
+  [#16863](https://github.com/google-gemini/gemini-cli/pull/16863)
+- feat: Add flash lite utility fallback chain by @adamfweidman in
+  [#17056](https://github.com/google-gemini/gemini-cli/pull/17056)
+- Fixes Windows crash: "Cannot resize a pty that has already exited" by @dzammit
+  in [#15757](https://github.com/google-gemini/gemini-cli/pull/15757)
+- feat(core): Add initial eval for generalist agent. by @joshualitt in
+  [#16856](https://github.com/google-gemini/gemini-cli/pull/16856)
+- feat(core): unify agent enabled and disabled flags by @SandyTao520 in
+  [#17127](https://github.com/google-gemini/gemini-cli/pull/17127)
+- fix(core): resolve auto model in default strategy by @sehoon38 in
+  [#17116](https://github.com/google-gemini/gemini-cli/pull/17116)
+- docs: update project context and pr-creator workflow by @NTaylorMullen in
+  [#17119](https://github.com/google-gemini/gemini-cli/pull/17119)
+- fix(cli): send gemini-cli version as mcp client version by @dsp in
+  [#13407](https://github.com/google-gemini/gemini-cli/pull/13407)
+- fix(cli): resolve Ctrl+Enter and Ctrl+J newline issues by @imadraude in
+  [#17021](https://github.com/google-gemini/gemini-cli/pull/17021)
+- Remove missing sidebar item by @chrstnb in
+  [#17145](https://github.com/google-gemini/gemini-cli/pull/17145)
+- feat(core): Ensure all properties in hooks object are event names. by
+  @joshualitt in
+  [#16870](https://github.com/google-gemini/gemini-cli/pull/16870)
+- fix(cli): fix newline support broken in previous PR by @scidomino in
+  [#17159](https://github.com/google-gemini/gemini-cli/pull/17159)
+- Add interactive ValidationDialog for handling 403 VALIDATION_REQUIRED errors.
+  by @gsquared94 in
+  [#16231](https://github.com/google-gemini/gemini-cli/pull/16231)
+- Add Esc-Esc to clear prompt when it's not empty by @Adib234 in
+  [#17131](https://github.com/google-gemini/gemini-cli/pull/17131)
+- Avoid spurious warnings about unexpected renders triggered by appEvents and
+  coreEvents. by @jacob314 in
+  [#17160](https://github.com/google-gemini/gemini-cli/pull/17160)
+- fix(cli): resolve home/end keybinding conflict by @scidomino in
+  [#17124](https://github.com/google-gemini/gemini-cli/pull/17124)
+- fix(cli): display 'http' type on mcp list by @pamanta in
+  [#16915](https://github.com/google-gemini/gemini-cli/pull/16915)
+- fix bad fallback logic external editor logic by @scidomino in
+  [#17166](https://github.com/google-gemini/gemini-cli/pull/17166)
+- Fix bug where System scopes weren't migrated. by @jacob314 in
+  [#17174](https://github.com/google-gemini/gemini-cli/pull/17174)
+- Fix mcp tool lookup in tool registry by @werdnum in
+  [#17054](https://github.com/google-gemini/gemini-cli/pull/17054)
 
-**Full Changelog**:
-https://github.com/google-gemini/gemini-cli/compare/v0.22.0-preview.3...v0.23.0-preview.0
+**Full changelog**:
+https://github.com/google-gemini/gemini-cli/compare/v0.25.0-preview.4...v0.26.0-preview.0

docs/cli/commands.md

@@ -73,6 +73,9 @@ Slash commands provide meta-level control over the CLI itself.
 - **`/copy`**
   - **Description:** Copies the last output produced by Gemini CLI to your
     clipboard, for easy sharing or reuse.
+  - **Behavior:**
+    - Local sessions use system clipboard tools (pbcopy/xclip/clip).
+    - Remote sessions (SSH/WSL) use OSC 52 and require terminal support.
   - **Note:** This command requires platform-specific clipboard tools to be
     installed.
     - On Linux, it requires `xclip` or `xsel`. You can typically install them
@@ -164,6 +167,7 @@ Slash commands provide meta-level control over the CLI itself.
   - **Note:** Only available if checkpointing is configured via
     [settings](../get-started/configuration.md). See
     [Checkpointing documentation](../cli/checkpointing.md) for more details.
+
 - **`/resume`**
   - **Description:** Browse and resume previous conversation sessions. Opens an
     interactive session browser where you can search, filter, and select from
@@ -194,6 +198,23 @@ Slash commands provide meta-level control over the CLI itself.
     modify them as desired. Changes to some settings are applied immediately,
     while others require a restart.
 
+- [**`/skills`**](./skills.md)
+  - **Description:** (Experimental) Manage Agent Skills, which provide on-demand
+    expertise and specialized workflows.
+  - **Sub-commands:**
+    - **`list`**:
+      - **Description:** List all discovered skills and their current status
+        (enabled/disabled).
+    - **`enable`**:
+      - **Description:** Enable a specific skill by name.
+      - **Usage:** `/skills enable <name>`
+    - **`disable`**:
+      - **Description:** Disable a specific skill by name.
+      - **Usage:** `/skills disable <name>`
+    - **`reload`**:
+      - **Description:** Refresh the list of discovered skills from all tiers
+        (workspace, user, and extensions).
+
 - **`/stats`**
   - **Description:** Display detailed statistics for the current Gemini CLI
     session, including token usage, cached token savings (when available), and

docs/cli/custom-commands.md

@@ -50,7 +50,7 @@ Your command definition files must be written in the TOML format and use the
 ## Handling arguments
 
 Custom commands support two powerful methods for handling arguments. The CLI
-automatically chooses the correct method based on the content of your command\'s
+automatically chooses the correct method based on the content of your command's
 `prompt`.
 
 ### 1. Context-aware injection with `{{args}}`
@@ -96,13 +96,13 @@ Search Results:
 """
 ```
 
-When you run `/grep-code It\'s complicated`:
+When you run `/grep-code It's complicated`:
 
 1. The CLI sees `{{args}}` used both outside and inside `!{...}`.
-2. Outside: The first `{{args}}` is replaced raw with `It\'s complicated`.
+2. Outside: The first `{{args}}` is replaced raw with `It's complicated`.
 3. Inside: The second `{{args}}` is replaced with the escaped version (e.g., on
    Linux: `"It\'s complicated"`).
-4. The command executed is `grep -r "It\'s complicated" .`.
+4. The command executed is `grep -r "It's complicated" .`.
 5. The CLI prompts you to confirm this exact, secure command before execution.
 6. The final prompt is sent.
 
@@ -129,13 +129,13 @@ format and behavior.
 # In: <project>/.gemini/commands/changelog.toml
 # Invoked via: /changelog 1.2.0 added "Support for default argument parsing."
 
-description = "Adds a new entry to the project\'s CHANGELOG.md file."
+description = "Adds a new entry to the project's CHANGELOG.md file."
 prompt = """
 # Task: Update Changelog
 
 You are an expert maintainer of this software project. A user has invoked a command to add a new entry to the changelog.
 
-**The user\'s raw command is appended below your instructions.**
+**The user's raw command is appended below your instructions.**
 
 Your task is to parse the `<version>`, `<change_type>`, and `<message>` from their input and use the `write_file` tool to correctly update the `CHANGELOG.md` file.
 
@@ -147,7 +147,7 @@ The command follows this format: `/changelog <version> <type> <message>`
 1. Read the `CHANGELOG.md` file.
 2. Find the section for the specified `<version>`.
 3. Add the `<message>` under the correct `<type>` heading.
-4. If the version or type section doesn\'t exist, create it.
+4. If the version or type section doesn't exist, create it.
 5. Adhere strictly to the "Keep a Changelog" format.
 """
 ```
@@ -241,7 +241,7 @@ operate on specific files.
 **Example (`review.toml`):**
 
 This command injects the content of a _fixed_ best practices file
-(`docs/best-practices.md`) and uses the user\'s arguments to provide context for
+(`docs/best-practices.md`) and uses the user's arguments to provide context for
 the review.
 
 ```toml
@@ -293,7 +293,7 @@ practice.
 description = "Asks the model to refactor the current context into a pure function."
 
 prompt = """
-Please analyze the code I\'ve provided in the current context.
+Please analyze the code I've provided in the current context.
 Refactor it into a pure function.
 
 Your response should include:

docs/cli/index.md

@@ -25,10 +25,12 @@ overview of Gemini CLI, see the [main documentation page](../index.md).
 
 - **[Checkpointing](./checkpointing.md):** Automatically save and restore
   snapshots of your session and files.
-- **[Enterprise configuration](./enterprise.md):** Deploying and manage Gemini
-  CLI in an enterprise environment.
+- **[Enterprise configuration](./enterprise.md):** Deploy and manage Gemini CLI
+  in an enterprise environment.
 - **[Sandboxing](./sandbox.md):** Isolate tool execution in a secure,
   containerized environment.
+- **[Agent Skills](./skills.md):** (Experimental) Extend the CLI with
+  specialized expertise and procedural workflows.
 - **[Telemetry](./telemetry.md):** Configure observability to monitor usage and
   performance.
 - **[Token caching](./token-caching.md):** Optimize API costs by caching tokens.

docs/cli/keyboard-shortcuts.md

@@ -8,43 +8,50 @@ available combinations.
 
 #### Basic Controls
 
-| Action                                       | Keys    |
-| -------------------------------------------- | ------- |
-| Confirm the current selection or choice.     | `Enter` |
-| Dismiss dialogs or cancel the current focus. | `Esc`   |
+| Action                                                          | Keys       |
+| --------------------------------------------------------------- | ---------- |
+| Confirm the current selection or choice.                        | `Enter`    |
+| Dismiss dialogs or cancel the current focus.                    | `Esc`      |
+| Cancel the current request or quit the CLI when input is empty. | `Ctrl + C` |
+| Exit the CLI when the input buffer is empty.                    | `Ctrl + D` |
 
 #### Cursor Movement
 
-| Action                                    | Keys                   |
-| ----------------------------------------- | ---------------------- |
-| Move the cursor to the start of the line. | `Ctrl + A`<br />`Home` |
-| Move the cursor to the end of the line.   | `Ctrl + E`<br />`End`  |
+| Action                                      | Keys                                                         |
+| ------------------------------------------- | ------------------------------------------------------------ |
+| Move the cursor to the start of the line.   | `Ctrl + A`<br />`Home (no Shift, Ctrl)`                      |
+| Move the cursor to the end of the line.     | `Ctrl + E`<br />`End (no Shift, Ctrl)`                       |
+| Move the cursor up one line.                | `Up Arrow (no Shift, Alt, Ctrl, Cmd)`                        |
+| Move the cursor down one line.              | `Down Arrow (no Shift, Alt, Ctrl, Cmd)`                      |
+| Move the cursor one character to the left.  | `Left Arrow (no Shift, Alt, Ctrl, Cmd)`<br />`Ctrl + B`      |
+| Move the cursor one character to the right. | `Right Arrow (no Shift, Alt, Ctrl, Cmd)`<br />`Ctrl + F`     |
+| Move the cursor one word to the left.       | `Ctrl + Left Arrow`<br />`Alt + Left Arrow`<br />`Alt + B`   |
+| Move the cursor one word to the right.      | `Ctrl + Right Arrow`<br />`Alt + Right Arrow`<br />`Alt + F` |
 
 #### Editing
 
-| Action                                           | Keys                                      |
-| ------------------------------------------------ | ----------------------------------------- |
-| Delete from the cursor to the end of the line.   | `Ctrl + K`                                |
-| Delete from the cursor to the start of the line. | `Ctrl + U`                                |
-| Clear all text in the input field.               | `Ctrl + C`                                |
-| Delete the previous word.                        | `Ctrl + Backspace`<br />`Cmd + Backspace` |
-
-#### Screen Control
-
-| Action                                       | Keys       |
-| -------------------------------------------- | ---------- |
-| Clear the terminal screen and redraw the UI. | `Ctrl + L` |
+| Action                                           | Keys                                                      |
+| ------------------------------------------------ | --------------------------------------------------------- |
+| Delete from the cursor to the end of the line.   | `Ctrl + K`                                                |
+| Delete from the cursor to the start of the line. | `Ctrl + U`                                                |
+| Clear all text in the input field.               | `Ctrl + C`                                                |
+| Delete the previous word.                        | `Ctrl + Backspace`<br />`Alt + Backspace`<br />`Ctrl + W` |
+| Delete the next word.                            | `Ctrl + Delete`<br />`Alt + Delete`                       |
+| Delete the character to the left.                | `Backspace`<br />`Ctrl + H`                               |
+| Delete the character to the right.               | `Delete`<br />`Ctrl + D`                                  |
+| Undo the most recent text edit.                  | `Ctrl + Z (no Shift)`                                     |
+| Redo the most recent undone text edit.           | `Shift + Ctrl + Z`                                        |
 
 #### Scrolling
 
-| Action                   | Keys                 |
-| ------------------------ | -------------------- |
-| Scroll content up.       | `Shift + Up Arrow`   |
-| Scroll content down.     | `Shift + Down Arrow` |
-| Scroll to the top.       | `Home`               |
-| Scroll to the bottom.    | `End`                |
-| Scroll up by one page.   | `Page Up`            |
-| Scroll down by one page. | `Page Down`          |
+| Action                   | Keys                              |
+| ------------------------ | --------------------------------- |
+| Scroll content up.       | `Shift + Up Arrow`                |
+| Scroll content down.     | `Shift + Down Arrow`              |
+| Scroll to the top.       | `Ctrl + Home`<br />`Shift + Home` |
+| Scroll to the bottom.    | `Ctrl + End`<br />`Shift + End`   |
+| Scroll up by one page.   | `Page Up`                         |
+| Scroll down by one page. | `Page Down`                       |
 
 #### History & Search
 
@@ -53,8 +60,9 @@ available combinations.
 | Show the previous entry in history.          | `Ctrl + P (no Shift)` |
 | Show the next entry in history.              | `Ctrl + N (no Shift)` |
 | Start reverse search through history.        | `Ctrl + R`            |
-| Insert the selected reverse-search match.    | `Enter (no Ctrl)`     |
+| Submit the selected reverse-search match.    | `Enter (no Ctrl)`     |
 | Accept a suggestion while reverse searching. | `Tab`                 |
+| Browse and rewind previous interactions.     | `Double Esc`          |
 
 #### Navigation
 
@@ -77,67 +85,45 @@ available combinations.
 
 #### Text Input
 
-| Action                               | Keys                                                                                        |
-| ------------------------------------ | ------------------------------------------------------------------------------------------- |
-| Submit the current prompt.           | `Enter (no Ctrl, no Shift, no Cmd, not Paste)`                                              |
-| Insert a newline without submitting. | `Ctrl + Enter`<br />`Cmd + Enter`<br />`Paste + Enter`<br />`Shift + Enter`<br />`Ctrl + J` |
-
-#### External Tools
-
-| Action                                         | Keys                      |
-| ---------------------------------------------- | ------------------------- |
-| Open the current prompt in an external editor. | `Ctrl + X`                |
-| Paste from the clipboard.                      | `Ctrl + V`<br />`Cmd + V` |
+| Action                                         | Keys                                                                                      |
+| ---------------------------------------------- | ----------------------------------------------------------------------------------------- |
+| Submit the current prompt.                     | `Enter (no Shift, Alt, Ctrl, Cmd)`                                                        |
+| Insert a newline without submitting.           | `Ctrl + Enter`<br />`Cmd + Enter`<br />`Alt + Enter`<br />`Shift + Enter`<br />`Ctrl + J` |
+| Open the current prompt in an external editor. | `Ctrl + X`                                                                                |
+| Paste from the clipboard.                      | `Ctrl + V`<br />`Cmd + V`<br />`Alt + V`                                                  |
 
 #### App Controls
 
-| Action                                                            | Keys       |
-| ----------------------------------------------------------------- | ---------- |
-| Toggle detailed error information.                                | `F12`      |
-| Toggle the full TODO list.                                        | `Ctrl + T` |
-| Toggle IDE context details.                                       | `Ctrl + G` |
-| Toggle Markdown rendering.                                        | `Cmd + M`  |
-| Toggle copy mode when the terminal is using the alternate buffer. | `Ctrl + S` |
-| Expand a height-constrained response to show additional lines.    | `Ctrl + S` |
-| Toggle focus between the shell and Gemini input.                  | `Ctrl + F` |
-
-#### Session Control
-
-| Action                                       | Keys       |
-| -------------------------------------------- | ---------- |
-| Cancel the current request or quit the CLI.  | `Ctrl + C` |
-| Exit the CLI when the input buffer is empty. | `Ctrl + D` |
+| Action                                                                                                | Keys                       |
+| ----------------------------------------------------------------------------------------------------- | -------------------------- |
+| Toggle detailed error information.                                                                    | `F12`                      |
+| Toggle the full TODO list.                                                                            | `Ctrl + T`                 |
+| Show IDE context details.                                                                             | `Ctrl + G`                 |
+| Toggle Markdown rendering.                                                                            | `Alt + M`                  |
+| Toggle copy mode when in alternate buffer mode.                                                       | `Ctrl + S`                 |
+| Toggle YOLO (auto-approval) mode for tool calls.                                                      | `Ctrl + Y`                 |
+| Cycle through approval modes: default (prompt), auto_edit (auto-approve edits), and plan (read-only). | `Shift + Tab`              |
+| Expand a height-constrained response to show additional lines when not in alternate buffer mode.      | `Ctrl + O`<br />`Ctrl + S` |
+| Focus the shell input from the gemini input.                                                          | `Tab (no Shift)`           |
+| Focus the Gemini input from the shell input.                                                          | `Tab`                      |
+| Clear the terminal screen and redraw the UI.                                                          | `Ctrl + L`                 |
+| Restart the application.                                                                              | `R`                        |
 
 <!-- KEYBINDINGS-AUTOGEN:END -->
 
 ## Additional context-specific shortcuts
 
-- `Ctrl+Y`: Toggle YOLO (auto-approval) mode for tool calls.
-- `Shift+Tab`: Toggle Auto Edit (auto-accept edits) mode.
-- `Option+M` (macOS): Entering `µ` with Option+M also toggles Markdown
-  rendering, matching `Cmd+M`.
+- `Option+B/F/M` (macOS only): Are interpreted as `Cmd+B/F/M` even if your
+  terminal isn't configured to send Meta with Option.
 - `!` on an empty prompt: Enter or exit shell mode.
 - `\` (at end of a line) + `Enter`: Insert a newline without leaving single-line
   mode.
-- `Ctrl+Delete` / `Meta+Delete`: Delete the word to the right of the cursor.
-- `Ctrl+B` or `Left Arrow`: Move the cursor one character to the left while
-  editing text.
-- `Ctrl+F` or `Right Arrow`: Move the cursor one character to the right; with an
-  embedded shell attached, `Ctrl+F` still toggles focus.
-- `Ctrl+D` or `Delete`: Remove the character immediately to the right of the
-  cursor.
-- `Ctrl+H` or `Backspace`: Remove the character immediately to the left of the
-  cursor.
-- `Ctrl+Left Arrow` / `Meta+Left Arrow` / `Meta+B`: Move one word to the left.
-- `Ctrl+Right Arrow` / `Meta+Right Arrow` / `Meta+F`: Move one word to the
-  right.
-- `Ctrl+W`: Delete the word to the left of the cursor (in addition to
-  `Ctrl+Backspace` / `Cmd+Backspace`).
-- `Ctrl+Z` / `Ctrl+Shift+Z`: Undo or redo the most recent text edit.
-- `Meta+Enter`: Open the current input in an external editor (alias for
-  `Ctrl+X`).
-- `Esc` pressed twice quickly: Clear the current input buffer.
+- `Esc` pressed twice quickly: Clear the input prompt if it is not empty,
+  otherwise browse and rewind previous interactions.
 - `Up Arrow` / `Down Arrow`: When the cursor is at the top or bottom of a
   single-line input, navigate backward or forward through prompt history.
 - `Number keys (1-9, multi-digit)` inside selection dialogs: Jump directly to
   the numbered radio option and confirm when the full number is entered.
+- `Double-click` on a paste placeholder (`[Pasted Text: X lines]`) in alternate
+  buffer mode: Expand to view full content inline. Double-click again to
+  collapse.

docs/cli/model-routing.md

@@ -11,12 +11,17 @@ health and automatically routes requests to available models based on defined
 policies.
 
 1.  **Model failure:** If the currently selected model fails (e.g., due to quota
-    or server errors), the CLI will iniate the fallback process.
+    or server errors), the CLI will initiate the fallback process.
 
 2.  **User consent:** Depending on the failure and the model's policy, the CLI
     may prompt you to switch to a fallback model (by default always prompts
     you).
 
+    Some internal utility calls (such as prompt completion and classification)
+    use a silent fallback chain for `gemini-2.5-flash-lite` and will fall back
+    to `gemini-2.5-flash` and `gemini-2.5-pro` without prompting or changing the
+    configured model.
+
 3.  **Model switch:** If approved, or if the policy allows for silent fallback,
     the CLI will use an available fallback model for the current turn or the
     remainder of the session.

docs/cli/sandbox.md

@@ -11,7 +11,7 @@ Before using sandboxing, you need to install and set up the Gemini CLI:
 npm install -g @google/gemini-cli
 ```
 
-To verify the installation
+To verify the installation:
 
 ```bash
 gemini --version

docs/cli/settings.md

@@ -18,45 +18,51 @@ Note: Workspace settings override user settings.
 Here is a list of all the available settings, grouped by category and ordered as
 they appear in the UI.
 
+<!-- SETTINGS-AUTOGEN:START -->
+
 ### General
 
-| UI Label                        | Setting                            | Description                                                                  | Default     |
-| ------------------------------- | ---------------------------------- | ---------------------------------------------------------------------------- | ----------- |
-| Preview Features (e.g., models) | `general.previewFeatures`          | Enable preview features (e.g., preview models).                              | `false`     |
-| Vim Mode                        | `general.vimMode`                  | Enable Vim keybindings.                                                      | `false`     |
-| Disable Auto Update             | `general.disableAutoUpdate`        | Disable automatic updates.                                                   | `false`     |
-| Enable Prompt Completion        | `general.enablePromptCompletion`   | Enable AI-powered prompt completion suggestions while typing.                | `false`     |
-| Debug Keystroke Logging         | `general.debugKeystrokeLogging`    | Enable debug logging of keystrokes to the console.                           | `false`     |
-| Session Retention               | `general.sessionRetention`         | Settings for automatic session cleanup. This feature is disabled by default. | `undefined` |
-| Enable Session Cleanup          | `general.sessionRetention.enabled` | Enable automatic session cleanup.                                            | `false`     |
+| UI Label                        | Setting                            | Description                                                   | Default |
+| ------------------------------- | ---------------------------------- | ------------------------------------------------------------- | ------- |
+| Preview Features (e.g., models) | `general.previewFeatures`          | Enable preview features (e.g., preview models).               | `false` |
+| Vim Mode                        | `general.vimMode`                  | Enable Vim keybindings                                        | `false` |
+| Enable Auto Update              | `general.enableAutoUpdate`         | Enable automatic updates.                                     | `true`  |
+| Enable Prompt Completion        | `general.enablePromptCompletion`   | Enable AI-powered prompt completion suggestions while typing. | `false` |
+| Debug Keystroke Logging         | `general.debugKeystrokeLogging`    | Enable debug logging of keystrokes to the console.            | `false` |
+| Enable Session Cleanup          | `general.sessionRetention.enabled` | Enable automatic session cleanup                              | `false` |
 
 ### Output
 
-| UI Label      | Setting         | Description                                            | Default |
-| ------------- | --------------- | ------------------------------------------------------ | ------- |
-| Output Format | `output.format` | The format of the CLI output. Can be `text` or `json`. | `text`  |
+| UI Label      | Setting         | Description                                            | Default  |
+| ------------- | --------------- | ------------------------------------------------------ | -------- |
+| Output Format | `output.format` | The format of the CLI output. Can be `text` or `json`. | `"text"` |
 
 ### UI
 
-| UI Label                       | Setting                                  | Description                                                          | Default |
-| ------------------------------ | ---------------------------------------- | -------------------------------------------------------------------- | ------- |
-| Hide Window Title              | `ui.hideWindowTitle`                     | Hide the window title bar.                                           | `false` |
-| Show Status in Title           | `ui.showStatusInTitle`                   | Show Gemini CLI status and thoughts in the terminal window title.    | `false` |
-| Hide Tips                      | `ui.hideTips`                            | Hide helpful tips in the UI.                                         | `false` |
-| Hide Banner                    | `ui.hideBanner`                          | Hide the application banner.                                         | `false` |
-| Hide Context Summary           | `ui.hideContextSummary`                  | Hide the context summary (GEMINI.md, MCP servers) above the input.   | `false` |
-| Hide CWD                       | `ui.footer.hideCWD`                      | Hide the current working directory path in the footer.               | `false` |
-| Hide Sandbox Status            | `ui.footer.hideSandboxStatus`            | Hide the sandbox status indicator in the footer.                     | `false` |
-| Hide Model Info                | `ui.footer.hideModelInfo`                | Hide the model name and context usage in the footer.                 | `false` |
-| Hide Context Window Percentage | `ui.footer.hideContextPercentage`        | Hides the context window remaining percentage.                       | `true`  |
-| Hide Footer                    | `ui.hideFooter`                          | Hide the footer from the UI.                                         | `false` |
-| Show Memory Usage              | `ui.showMemoryUsage`                     | Display memory usage information in the UI.                          | `false` |
-| Show Line Numbers              | `ui.showLineNumbers`                     | Show line numbers in the chat.                                       | `false` |
-| Show Citations                 | `ui.showCitations`                       | Show citations for generated text in the chat.                       | `false` |
-| Use Full Width                 | `ui.useFullWidth`                        | Use the entire width of the terminal for output.                     | `true`  |
-| Use Alternate Screen Buffer    | `ui.useAlternateBuffer`                  | Use an alternate screen buffer for the UI, preserving shell history. | `true`  |
-| Disable Loading Phrases        | `ui.accessibility.disableLoadingPhrases` | Disable loading phrases for accessibility.                           | `false` |
-| Screen Reader Mode             | `ui.accessibility.screenReader`          | Render output in plain-text to be more screen reader accessible.     | `false` |
+| UI Label                       | Setting                                 | Description                                                                                                                                                       | Default |
+| ------------------------------ | --------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------- |
+| Hide Window Title              | `ui.hideWindowTitle`                    | Hide the window title bar                                                                                                                                         | `false` |
+| Show Thoughts in Title         | `ui.showStatusInTitle`                  | Show Gemini CLI model thoughts in the terminal window title during the working phase                                                                              | `false` |
+| Dynamic Window Title           | `ui.dynamicWindowTitle`                 | Update the terminal window title with current status icons (Ready: ◇, Action Required: ✋, Working: ✦)                                                            | `true`  |
+| Show Home Directory Warning    | `ui.showHomeDirectoryWarning`           | Show a warning when running Gemini CLI in the home directory.                                                                                                     | `true`  |
+| Hide Tips                      | `ui.hideTips`                           | Hide helpful tips in the UI                                                                                                                                       | `false` |
+| Hide Banner                    | `ui.hideBanner`                         | Hide the application banner                                                                                                                                       | `false` |
+| Hide Context Summary           | `ui.hideContextSummary`                 | Hide the context summary (GEMINI.md, MCP servers) above the input.                                                                                                | `false` |
+| Hide CWD                       | `ui.footer.hideCWD`                     | Hide the current working directory path in the footer.                                                                                                            | `false` |
+| Hide Sandbox Status            | `ui.footer.hideSandboxStatus`           | Hide the sandbox status indicator in the footer.                                                                                                                  | `false` |
+| Hide Model Info                | `ui.footer.hideModelInfo`               | Hide the model name and context usage in the footer.                                                                                                              | `false` |
+| Hide Context Window Percentage | `ui.footer.hideContextPercentage`       | Hides the context window remaining percentage.                                                                                                                    | `true`  |
+| Hide Footer                    | `ui.hideFooter`                         | Hide the footer from the UI                                                                                                                                       | `false` |
+| Show Memory Usage              | `ui.showMemoryUsage`                    | Display memory usage information in the UI                                                                                                                        | `false` |
+| Show Line Numbers              | `ui.showLineNumbers`                    | Show line numbers in the chat.                                                                                                                                    | `true`  |
+| Show Citations                 | `ui.showCitations`                      | Show citations for generated text in the chat.                                                                                                                    | `false` |
+| Show Model Info In Chat        | `ui.showModelInfoInChat`                | Show the model name in the chat for each model turn.                                                                                                              | `false` |
+| Use Alternate Screen Buffer    | `ui.useAlternateBuffer`                 | Use an alternate screen buffer for the UI, preserving shell history.                                                                                              | `false` |
+| Use Background Color           | `ui.useBackgroundColor`                 | Whether to use background colors in the UI.                                                                                                                       | `true`  |
+| Incremental Rendering          | `ui.incrementalRendering`               | Enable incremental rendering for the UI. This option will reduce flickering but may cause rendering artifacts. Only supported when useAlternateBuffer is enabled. | `true`  |
+| Show Spinner                   | `ui.showSpinner`                        | Show the spinner during operations.                                                                                                                               | `true`  |
+| Enable Loading Phrases         | `ui.accessibility.enableLoadingPhrases` | Enable loading phrases during operations.                                                                                                                         | `true`  |
+| Screen Reader Mode             | `ui.accessibility.screenReader`         | Render output in plain-text to be more screen reader accessible                                                                                                   | `false` |
 
 ### IDE
 
@@ -69,7 +75,7 @@ they appear in the UI.
 | UI Label                | Setting                      | Description                                                                            | Default |
 | ----------------------- | ---------------------------- | -------------------------------------------------------------------------------------- | ------- |
 | Max Session Turns       | `model.maxSessionTurns`      | Maximum number of user/model/tool turns to keep in a session. -1 means unlimited.      | `-1`    |
-| Compression Threshold   | `model.compressionThreshold` | The fraction of context usage at which to trigger context compression (e.g. 0.2, 0.3). | `0.2`   |
+| Compression Threshold   | `model.compressionThreshold` | The fraction of context usage at which to trigger context compression (e.g. 0.2, 0.3). | `0.5`   |
 | Skip Next Speaker Check | `model.skipNextSpeakerCheck` | Skip the next speaker check.                                                           | `true`  |
 
 ### Context
@@ -81,35 +87,44 @@ they appear in the UI.
 | Respect .gitignore                   | `context.fileFiltering.respectGitIgnore`          | Respect .gitignore files when searching.                                                                                                        | `true`  |
 | Respect .geminiignore                | `context.fileFiltering.respectGeminiIgnore`       | Respect .geminiignore files when searching.                                                                                                     | `true`  |
 | Enable Recursive File Search         | `context.fileFiltering.enableRecursiveFileSearch` | Enable recursive file search functionality when completing @ references in the prompt.                                                          | `true`  |
-| Disable Fuzzy Search                 | `context.fileFiltering.disableFuzzySearch`        | Disable fuzzy search when searching for files.                                                                                                  | `false` |
+| Enable Fuzzy Search                  | `context.fileFiltering.enableFuzzySearch`         | Enable fuzzy search when searching for files.                                                                                                   | `true`  |
 
 ### Tools
 
-| UI Label                         | Setting                              | Description                                                                                                     | Default |
-| -------------------------------- | ------------------------------------ | --------------------------------------------------------------------------------------------------------------- | ------- |
-| Enable Interactive Shell         | `tools.shell.enableInteractiveShell` | Use node-pty for an interactive shell experience. Fallback to child_process still applies.                      | `true`  |
-| Show Color                       | `tools.shell.showColor`              | Show color in shell output.                                                                                     | `false` |
-| Auto Accept                      | `tools.autoAccept`                   | Automatically accept and execute tool calls that are considered safe (e.g., read-only operations).              | `false` |
-| Use Ripgrep                      | `tools.useRipgrep`                   | Use ripgrep for file content search instead of the fallback implementation. Provides faster search performance. | `true`  |
-| Enable Tool Output Truncation    | `tools.enableToolOutputTruncation`   | Enable truncation of large tool outputs.                                                                        | `true`  |
-| Tool Output Truncation Threshold | `tools.truncateToolOutputThreshold`  | Truncate tool output if it is larger than this many characters. Set to -1 to disable.                           | `10000` |
-| Tool Output Truncation Lines     | `tools.truncateToolOutputLines`      | The number of lines to keep when truncating tool output.                                                        | `100`   |
+| UI Label                         | Setting                              | Description                                                                                                                                                                    | Default     |
+| -------------------------------- | ------------------------------------ | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | ----------- |
+| Enable Interactive Shell         | `tools.shell.enableInteractiveShell` | Use node-pty for an interactive shell experience. Fallback to child_process still applies.                                                                                     | `true`      |
+| Show Color                       | `tools.shell.showColor`              | Show color in shell output.                                                                                                                                                    | `false`     |
+| Auto Accept                      | `tools.autoAccept`                   | Automatically accept and execute tool calls that are considered safe (e.g., read-only operations).                                                                             | `false`     |
+| Approval Mode                    | `tools.approvalMode`                 | The default approval mode for tool execution. 'default' prompts for approval, 'auto_edit' auto-approves edit tools, and 'plan' is read-only mode. 'yolo' is not supported yet. | `"default"` |
+| Use Ripgrep                      | `tools.useRipgrep`                   | Use ripgrep for file content search instead of the fallback implementation. Provides faster search performance.                                                                | `true`      |
+| Enable Tool Output Truncation    | `tools.enableToolOutputTruncation`   | Enable truncation of large tool outputs.                                                                                                                                       | `true`      |
+| Tool Output Truncation Threshold | `tools.truncateToolOutputThreshold`  | Truncate tool output if it is larger than this many characters. Set to -1 to disable.                                                                                          | `4000000`   |
+| Tool Output Truncation Lines     | `tools.truncateToolOutputLines`      | The number of lines to keep when truncating tool output.                                                                                                                       | `1000`      |
+| Disable LLM Correction           | `tools.disableLLMCorrection`         | Disable LLM-based error correction for edit tools. When enabled, tools will fail immediately if exact string matches are not found, instead of attempting to self-correct.     | `true`      |
 
 ### Security
 
-| UI Label                      | Setting                                         | Description                                               | Default |
-| ----------------------------- | ----------------------------------------------- | --------------------------------------------------------- | ------- |
-| Disable YOLO Mode             | `security.disableYoloMode`                      | Disable YOLO mode, even if enabled by a flag.             | `false` |
-| Blocks extensions from Git    | `security.blockGitExtensions`                   | Blocks installing and loading extensions from Git.        | `false` |
-| Folder Trust                  | `security.folderTrust.enabled`                  | Setting to track whether Folder trust is enabled.         | `false` |
-| Allowed Environment Variables | `security.environmentVariableRedaction.allowed` | Environment variables to always allow (bypass redaction). | `[]`    |
-| Blocked Environment Variables | `security.environmentVariableRedaction.blocked` | Environment variables to always redact.                   | `[]`    |
+| UI Label                              | Setting                                         | Description                                                                     | Default |
+| ------------------------------------- | ----------------------------------------------- | ------------------------------------------------------------------------------- | ------- |
+| Disable YOLO Mode                     | `security.disableYoloMode`                      | Disable YOLO mode, even if enabled by a flag.                                   | `false` |
+| Allow Permanent Tool Approval         | `security.enablePermanentToolApproval`          | Enable the "Allow for all future sessions" option in tool confirmation dialogs. | `false` |
+| Blocks extensions from Git            | `security.blockGitExtensions`                   | Blocks installing and loading extensions from Git.                              | `false` |
+| Folder Trust                          | `security.folderTrust.enabled`                  | Setting to track whether Folder trust is enabled.                               | `false` |
+| Enable Environment Variable Redaction | `security.environmentVariableRedaction.enabled` | Enable redaction of environment variables that may contain secrets.             | `false` |
 
 ### Experimental
 
-| UI Label                            | Setting                                                 | Description                                                  | Default |
-| ----------------------------------- | ------------------------------------------------------- | ------------------------------------------------------------ | ------- |
-| Enable Codebase Investigator        | `experimental.codebaseInvestigatorSettings.enabled`     | Enable the Codebase Investigator agent.                      | `true`  |
-| Codebase Investigator Max Num Turns | `experimental.codebaseInvestigatorSettings.maxNumTurns` | Maximum number of turns for the Codebase Investigator agent. | `10`    |
-| Enable CLI Help Agent               | `experimental.cliHelpAgentSettings.enabled`             | Enable the CLI Help Agent.                                   | `true`  |
-| Agent Skills                        | `experimental.skills`                                   | Enable Agent Skills (experimental).                          | `false` |
+| UI Label         | Setting                      | Description                                                                         | Default |
+| ---------------- | ---------------------------- | ----------------------------------------------------------------------------------- | ------- |
+| Agent Skills     | `experimental.skills`        | Enable Agent Skills (experimental).                                                 | `false` |
+| Use OSC 52 Paste | `experimental.useOSC52Paste` | Use OSC 52 sequence for pasting instead of clipboardy (useful for remote sessions). | `false` |
+| Plan             | `experimental.plan`          | Enable planning features (Plan Mode and tools).                                     | `false` |
+
+### HooksConfig
+
+| UI Label           | Setting                     | Description                                      | Default |
+| ------------------ | --------------------------- | ------------------------------------------------ | ------- |
+| Hook Notifications | `hooksConfig.notifications` | Show visual indicators when hooks are executing. | `true`  |
+
+<!-- SETTINGS-AUTOGEN:END -->

docs/cli/skills.md

@@ -13,7 +13,7 @@ discoverable capability.
 ## Overview
 
 Unlike general context files ([`GEMINI.md`](./gemini-md.md)), which provide
-persistent project-wide background, Skills represent **on-demand expertise**.
+persistent workspace-wide background, Skills represent **on-demand expertise**.
 This allows Gemini to maintain a vast library of specialized capabilities—such
 as security auditing, cloud deployments, or codebase migrations—without
 cluttering the model's immediate context window.
@@ -39,15 +39,15 @@ the full instructions and resources required to complete the task using the
 
 Gemini CLI discovers skills from three primary locations:
 
-1.  **Project Skills** (`.gemini/skills/`): Project-specific skills that are
+1.  **Workspace Skills** (`.gemini/skills/`): Workspace-specific skills that are
     typically committed to version control and shared with the team.
 2.  **User Skills** (`~/.gemini/skills/`): Personal skills available across all
-    your projects.
+    your workspaces.
 3.  **Extension Skills**: Skills bundled within installed
     [extensions](../extensions/index.md).
 
 **Precedence:** If multiple skills share the same name, higher-precedence
-locations override lower ones: **Project > User > Extension**.
+locations override lower ones: **Workspace > User > Extension**.
 
 ## Managing Skills
 
@@ -61,7 +61,7 @@ Use the `/skills` slash command to view and manage available expertise:
 - `/skills reload`: Refreshes the list of discovered skills from all tiers.
 
 _Note: `/skills disable` and `/skills enable` default to the `user` scope. Use
-`--scope project` to manage project-specific settings._
+`--scope workspace` to manage workspace-specific settings._
 
 ### From the Terminal
 
@@ -71,9 +71,26 @@ The `gemini skills` command provides management utilities:
 # List all discovered skills
 gemini skills list
 
-# Enable/disable skills. Can use --scope to specify project or user
+# Install a skill from a Git repository, local directory, or zipped skill file (.skill)
+# Uses the user scope by default (~/.gemini/skills)
+gemini skills install https://github.com/user/repo.git
+gemini skills install /path/to/local/skill
+gemini skills install /path/to/local/my-expertise.skill
+
+# Install a specific skill from a monorepo or subdirectory using --path
+gemini skills install https://github.com/my-org/my-skills.git --path skills/frontend-design
+
+# Install to the workspace scope (.gemini/skills)
+gemini skills install /path/to/skill --scope workspace
+
+# Uninstall a skill by name
+gemini skills uninstall my-expertise --scope workspace
+
+# Enable a skill (globally)
 gemini skills enable my-expertise
-gemini skills disable my-expertise
+
+# Disable a skill. Can use --scope to specify workspace or user (defaults to workspace)
+gemini skills disable my-expertise --scope workspace
 ```
 
 ## Creating a Skill
@@ -81,7 +98,20 @@ gemini skills disable my-expertise
 A skill is a directory containing a `SKILL.md` file at its root. This file uses
 YAML frontmatter for metadata and Markdown for instructions.
 
-### Basic Structure
+### Folder Structure
+
+Skills are self-contained directories. At a minimum, a skill requires a
+`SKILL.md` file, but can include other resources:
+
+```text
+my-skill/
+├── SKILL.md       (Required) Instructions and metadata
+├── scripts/       (Optional) Executable scripts/tools
+├── references/    (Optional) Static documentation and examples
+└── assets/        (Optional) Templates and binary resources
+```
+
+### Basic Structure (SKILL.md)
 
 ```markdown
 ---
@@ -100,6 +130,8 @@ description: <what the skill does and when Gemini should use it>
 
 ### Example: Team Code Reviewer
 
+Create `~/.gemini/skills/code-reviewer/SKILL.md`:
+
 ```markdown
 ---
 name: code-reviewer
@@ -115,7 +147,7 @@ You are an expert code reviewer. When reviewing code, follow this workflow:
 1.  **Analyze**: Review the staged changes or specific files provided. Ensure
     that the changes are scoped properly and represent minimal changes required
     to address the issue.
-2.  **Style**: Ensure code follows the project's conventions and idiomatic
+2.  **Style**: Ensure code follows the workspace's conventions and idiomatic
     patterns as described in the `GEMINI.md` file.
 3.  **Security**: Flag any potential security vulnerabilities.
 4.  **Tests**: Verify that new logic has corresponding test coverage and that

docs/cli/system-prompt.md

@@ -56,6 +56,38 @@ error with: `missing system prompt file '<path>'`.
 When `GEMINI_SYSTEM_MD` is active, the CLI shows a `|⌐■_■|` indicator in the UI
 to signal custom system‑prompt mode.
 
+## Variable Substitution
+
+When using a custom system prompt file, you can use the following variables to
+dynamically include built-in content:
+
+- `${AgentSkills}`: Injects a complete section (including header) of all
+  available agent skills.
+- `${SubAgents}`: Injects a complete section (including header) of available
+  sub-agents.
+- `${AvailableTools}`: Injects a bulleted list of all currently enabled tool
+  names.
+- Tool Name Variables: Injects the actual name of a tool using the pattern:
+  `${toolName}_ToolName` (e.g., `${write_file_ToolName}`,
+  `${run_shell_command_ToolName}`).
+
+  This pattern is generated dynamically for all available tools.
+
+### Example
+
+```markdown
+# Custom System Prompt
+
+You are a helpful assistant. ${AgentSkills}
+${SubAgents}
+
+## Tooling
+
+The following tools are available to you: ${AvailableTools}
+
+You can use ${write_file_ToolName} to save logs.
+```
+
 ## Export the default prompt (recommended)
 
 Before overriding, export the current default prompt so you can review required

docs/cli/telemetry.md

@@ -8,14 +8,17 @@ Learn how to enable and setup OpenTelemetry for Gemini CLI.
   - [Configuration](#configuration)
   - [Google Cloud telemetry](#google-cloud-telemetry)
     - [Prerequisites](#prerequisites)
+    - [Authenticating with CLI Credentials](#authenticating-with-cli-credentials)
     - [Direct export (recommended)](#direct-export-recommended)
     - [Collector-based export (advanced)](#collector-based-export-advanced)
+    - [Monitoring Dashboards](#monitoring-dashboards)
   - [Local telemetry](#local-telemetry)
     - [File-based output (recommended)](#file-based-output-recommended)
     - [Collector-based export (advanced)](#collector-based-export-advanced-1)
   - [Logs and metrics](#logs-and-metrics)
     - [Logs](#logs)
       - [Sessions](#sessions)
+      - [Approval Mode](#approval-mode)
       - [Tools](#tools)
       - [Files](#files)
       - [API](#api)
@@ -213,6 +216,24 @@ forward data to Google Cloud.
    - Open `~/.gemini/tmp/<projectHash>/otel/collector-gcp.log` to view local
      collector logs.
 
+### Monitoring Dashboards
+
+Gemini CLI provides a pre-configured
+[Google Cloud Monitoring](https://cloud.google.com/monitoring) dashboard to
+visualize your telemetry.
+
+This dashboard can be found under **Google Cloud Monitoring Dashboard
+Templates** as "**Gemini CLI Monitoring**".
+
+![Gemini CLI Monitoring Dashboard Overview](/docs/assets/monitoring-dashboard-overview.png)
+
+![Gemini CLI Monitoring Dashboard Metrics](/docs/assets/monitoring-dashboard-metrics.png)
+
+![Gemini CLI Monitoring Dashboard Logs](/docs/assets/monitoring-dashboard-logs.png)
+
+To learn more, check out this blog post:
+[Instant insights: Gemini CLI’s new pre-configured monitoring dashboards](https://cloud.google.com/blog/topics/developers-practitioners/instant-insights-gemini-clis-new-pre-configured-monitoring-dashboards/).
+
 ## Local telemetry
 
 For local development and debugging, you can capture telemetry data locally:
@@ -295,6 +316,20 @@ Captures startup configuration and user prompt submissions.
     - `prompt` (string; excluded if `telemetry.logPrompts` is `false`)
     - `auth_type` (string)
 
+#### Approval Mode
+
+Tracks changes and duration of approval modes.
+
+- `approval_mode_switch`: Approval mode was changed.
+  - **Attributes**:
+    - `from_mode` (string)
+    - `to_mode` (string)
+
+- `approval_mode_duration`: Duration spent in an approval mode.
+  - **Attributes**:
+    - `mode` (string)
+    - `duration_ms` (int)
+
 #### Tools
 
 Captures tool executions, output truncation, and Edit behavior.

docs/cli/themes.md

@@ -86,7 +86,6 @@ color keys. For example:
 - `Gray`
 - `DiffAdded` (optional, for added lines in diffs)
 - `DiffRemoved` (optional, for removed lines in diffs)
-- `DiffModified` (optional, for modified lines in diffs)
 
 You can also override individual UI text roles by adding a nested `text` object.
 This object supports the keys `primary`, `secondary`, `link`, `accent`, and
@@ -157,7 +156,6 @@ custom theme defined in `settings.json`.
   "Gray": "#ABB2BF",
   "DiffAdded": "#A6E3A1",
   "DiffRemoved": "#F38BA8",
-  "DiffModified": "#89B4FA",
   "GradientColors": ["#4796E4", "#847ACE", "#C3677F"]
 }
 ```

docs/cli/uninstall.md

@@ -45,3 +45,21 @@ npm uninstall -g @google/gemini-cli
 ```
 
 This command completely removes the package from your system.
+
+## Method 3: Homebrew
+
+If you installed the CLI globally using Homebrew (e.g.,
+`brew install gemini-cli`), use the `brew uninstall` command to remove it.
+
+```bash
+brew uninstall gemini-cli
+```
+
+## Method 4: MacPorts
+
+If you installed the CLI globally using MacPorts (e.g.,
+`sudo port install gemini-cli`), use the `port uninstall` command to remove it.
+
+```bash
+sudo port uninstall gemini-cli
+```

docs/core/index.md

@@ -68,6 +68,10 @@ If you are using the default "pro" model and the CLI detects that you are being
 rate-limited, it automatically switches to the "flash" model for the current
 session. This allows you to continue working without interruption.
 
+Internal utility calls that use `gemini-2.5-flash-lite` (for example, prompt
+completion and classification) silently fall back to `gemini-2.5-flash` and
+`gemini-2.5-pro` when quota is exhausted, without changing the configured model.
+
 ## File discovery service
 
 The file discovery service is responsible for finding files in the project that

docs/core/memport.md

@@ -83,7 +83,9 @@ The processor automatically detects and prevents circular imports:
 # file-a.md
 
 @./file-b.md
+```
 
+```markdown
 # file-b.md
 
 @./file-a.md <!-- This will be detected and prevented -->

docs/core/policy-engine.md

@@ -146,6 +146,38 @@ A rule matches a tool call if all of its conditions are met:
 Policies are defined in `.toml` files. The CLI loads these files from Default,
 User, and (if configured) Admin directories.
 
+### Policy locations
+
+| Tier      | Type   | Location                    |
+| :-------- | :----- | :-------------------------- |
+| **User**  | Custom | `~/.gemini/policies/*.toml` |
+| **Admin** | System | _See below (OS specific)_   |
+
+#### System-wide policies (Admin)
+
+Administrators can enforce system-wide policies (Tier 3) that override all user
+and default settings. These policies must be placed in specific, secure
+directories:
+
+| OS          | Policy Directory Path                             |
+| :---------- | :------------------------------------------------ |
+| **Linux**   | `/etc/gemini-cli/policies`                        |
+| **macOS**   | `/Library/Application Support/GeminiCli/policies` |
+| **Windows** | `C:\ProgramData\gemini-cli\policies`              |
+
+**Security Requirements:**
+
+To prevent privilege escalation, the CLI enforces strict security checks on
+admin directories. If checks fail, system policies are **ignored**.
+
+- **Linux / macOS:** Must be owned by `root` (UID 0) and NOT writable by group
+  or others (e.g., `chmod 755`).
+- **Windows:** Must be in `C:\ProgramData`. Standard users (`Users`, `Everyone`)
+  must NOT have `Write`, `Modify`, or `Full Control` permissions. _Tip: If you
+  see a security warning, use the folder properties to remove write permissions
+  for non-admin groups. You may need to "Disable inheritance" in Advanced
+  Security Settings._
+
 ### TOML rule schema
 
 Here is a breakdown of the fields available in a TOML policy rule:
@@ -258,8 +290,9 @@ The Gemini CLI ships with a set of default policies to provide a safe
 out-of-the-box experience.
 
 - **Read-only tools** (like `read_file`, `glob`) are generally **allowed**.
-- **Agent delegation** (like `delegate_to_agent`) is **allowed** (sub-agent
-  actions are checked individually).
+- **Agent delegation** (like `delegate_to_agent`) defaults to **`ask_user`** to
+  ensure remote agents can prompt for confirmation, but local sub-agent actions
+  are executed silently and checked individually.
 - **Write tools** (like `write_file`, `run_shell_command`) default to
   **`ask_user`**.
 - In **`yolo`** mode, a high-priority rule allows all tools.

docs/extensions/best-practices.md

@@ -0,0 +1,139 @@
+# Extensions on Gemini CLI: Best practices
+
+This guide covers best practices for developing, securing, and maintaining
+Gemini CLI extensions.
+
+## Development
+
+Developing extensions for Gemini CLI is intended to be a lightweight, iterative
+process.
+
+### Structure your extension
+
+While simple extensions can just be a few files, we recommend a robust structure
+for complex extensions:
+
+```
+my-extension/
+├── package.json
+├── tsconfig.json
+├── gemini-extension.json
+├── src/
+│   ├── index.ts
+│   └── tools/
+└── dist/
+```
+
+- **Use TypeScript**: We strongly recommend using TypeScript for type safety and
+  better tooling.
+- **Separate source and build**: Keep your source code in `src` and build to
+  `dist`.
+- **Bundle dependencies**: If your extension has many dependencies, consider
+  bundling them (e.g., with `esbuild` or `webpack`) to reduce install time and
+  potential conflicts.
+
+### Iterate with `link`
+
+Use `gemini extensions link` to develop locally without constantly reinstalling:
+
+```bash
+cd my-extension
+gemini extensions link .
+```
+
+Changes to your code (after rebuilding) will be immediately available in the CLI
+on restart.
+
+### Use `GEMINI.md` effectively
+
+Your `GEMINI.md` file provides context to the model. Keep it focused:
+
+- **Do:** Explain high-level goals and how to use the provided tools.
+- **Don't:** Dump your entire documentation.
+- **Do:** Use clear, concise language.
+
+## Security
+
+When building a Gemini CLI extension, follow general security best practices
+(such as least privilege and input validation) to reduce risk.
+
+### Minimal permissions
+
+When defining tools in your MCP server, only request the permissions necessary.
+Avoid giving the model broad access (like full shell access) if a more
+restricted set of tools will suffice.
+
+If you must use powerful tools like `run_shell_command`, consider restricting
+them to specific commands in your `gemini-extension.json`:
+
+```json
+{
+  "name": "my-safe-extension",
+  "excludeTools": ["run_shell_command(rm -rf *)"]
+}
+```
+
+This ensures that even if the model tries to execute a dangerous command, it
+will be blocked at the CLI level.
+
+### Validate inputs
+
+Your MCP server is running on the user's machine. Always validate inputs to your
+tools to prevent arbitrary code execution or filesystem access outside the
+intended scope.
+
+```typescript
+// Good: Validating paths
+if (!path.resolve(inputPath).startsWith(path.resolve(allowedDir) + path.sep)) {
+  throw new Error('Access denied');
+}
+```
+
+### Sensitive settings
+
+If your extension requires API keys, use the `sensitive: true` option in
+`gemini-extension.json`. This ensures keys are stored securely in the system
+keychain and obfuscated in the UI.
+
+```json
+"settings": [
+  {
+    "name": "API Key",
+    "envVar": "MY_API_KEY",
+    "sensitive": true
+  }
+]
+```
+
+## Releasing
+
+You can upload your extension directly to GitHub to list it in the gallery.
+Gemini CLI extensions also offers support for more complicated
+[releases](releasing.md).
+
+### Semantic versioning
+
+Follow [Semantic Versioning](https://semver.org/).
+
+- **Major**: Breaking changes (renaming tools, changing arguments).
+- **Minor**: New features (new tools, commands).
+- **Patch**: Bug fixes.
+
+### Release Channels
+
+Use git branches to manage release channels (e.g., `main` for stable, `dev` for
+bleeding edge). This allows users to choose their stability level:
+
+```bash
+# Stable
+gemini extensions install github.com/user/repo
+
+# Dev
+gemini extensions install github.com/user/repo --ref dev
+```
+
+### Clean artifacts
+
+If you are using GitHub Releases, ensure your release artifacts only contain the
+necessary files (`dist/`, `gemini-extension.json`, `package.json`). Exclude
+`node_modules` (users will install them) and `src/` to keep downloads small.

docs/extensions/index.md

@@ -1,343 +1,44 @@
 # Gemini CLI extensions
 
-_This documentation is up-to-date with the v0.4.0 release._
-
-Gemini CLI extensions package prompts, MCP servers, and custom commands into a
-familiar and user-friendly format. With extensions, you can expand the
-capabilities of Gemini CLI and share those capabilities with others. They are
-designed to be easily installable and shareable.
+Gemini CLI extensions package prompts, MCP servers, custom commands, hooks, and
+agent skills into a familiar and user-friendly format. With extensions, you can
+expand the capabilities of Gemini CLI and share those capabilities with others.
+They are designed to be easily installable and shareable.
 
 To see examples of extensions, you can browse a gallery of
 [Gemini CLI extensions](https://geminicli.com/extensions/browse/).
 
-See [getting started docs](getting-started-extensions.md) for a guide on
-creating your first extension.
+## Managing extensions
 
-See [releasing docs](extension-releasing.md) for an advanced guide on setting up
-GitHub releases.
+You can verify your installed extensions and their status using the interactive
+command:
 
-## Extension management
-
-We offer a suite of extension management tools using `gemini extensions`
-commands.
-
-Note that these commands are not supported from within the CLI, although you can
-list installed extensions using the `/extensions list` subcommand.
-
-Note that all of these commands will only be reflected in active CLI sessions on
-restart.
-
-### Installing an extension
-
-You can install an extension using `gemini extensions install` with either a
-GitHub URL or a local path.
-
-Note that we create a copy of the installed extension, so you will need to run
-`gemini extensions update` to pull in changes from both locally-defined
-extensions and those on GitHub.
-
-NOTE: If you are installing an extension from GitHub, you'll need to have `git`
-installed on your machine. See
-[git installation instructions](https://git-scm.com/book/en/v2/Getting-Started-Installing-Git)
-for help.
-
-```
-gemini extensions install <source> [--ref <ref>] [--auto-update] [--pre-release] [--consent]
+```bash
+/extensions list
 ```
 
-- `<source>`: The github URL or local path of the extension to install.
-- `--ref`: The git ref to install from.
-- `--auto-update`: Enable auto-update for this extension.
-- `--pre-release`: Enable pre-release versions for this extension.
-- `--consent`: Acknowledge the security risks of installing an extension and
-  skip the confirmation prompt.
+or in noninteractive mode:
 
-### Uninstalling an extension
-
-To uninstall one or more extensions, run
-`gemini extensions uninstall <name...>`:
-
-```
-gemini extensions uninstall gemini-cli-security gemini-cli-another-extension
-```
-
-### Disabling an extension
-
-Extensions are, by default, enabled across all workspaces. You can disable an
-extension entirely or for specific workspace.
-
-```
-gemini extensions disable <name> [--scope <scope>]
-```
-
-- `<name>`: The name of the extension to disable.
-- `--scope`: The scope to disable the extension in (`user` or `workspace`).
-
-### Enabling an extension
-
-You can enable extensions using `gemini extensions enable <name>`. You can also
-enable an extension for a specific workspace using
-`gemini extensions enable <name> --scope=workspace` from within that workspace.
-
-```
-gemini extensions enable <name> [--scope <scope>]
-```
-
-- `<name>`: The name of the extension to enable.
-- `--scope`: The scope to enable the extension in (`user` or `workspace`).
-
-### Updating an extension
-
-For extensions installed from a local path or a git repository, you can
-explicitly update to the latest version (as reflected in the
-`gemini-extension.json` `version` field) with `gemini extensions update <name>`.
-
-You can update all extensions with:
-
-```
-gemini extensions update --all
-```
-
-### Create a boilerplate extension
-
-We offer several example extensions `context`, `custom-commands`,
-`exclude-tools` and `mcp-server`. You can view these examples
-[here](https://github.com/google-gemini/gemini-cli/tree/main/packages/cli/src/commands/extensions/examples).
-
-To copy one of these examples into a development directory using the type of
-your choosing, run:
-
-```
-gemini extensions new <path> [template]
-```
-
-- `<path>`: The path to create the extension in.
-- `[template]`: The boilerplate template to use.
-
-### Link a local extension
-
-The `gemini extensions link` command will create a symbolic link from the
-extension installation directory to the development path.
-
-This is useful so you don't have to run `gemini extensions update` every time
-you make changes you'd like to test.
-
-```
-gemini extensions link <path>
-```
-
-- `<path>`: The path of the extension to link.
-
-## How it works
-
-On startup, Gemini CLI looks for extensions in `<home>/.gemini/extensions`
-
-Extensions exist as a directory that contains a `gemini-extension.json` file.
-For example:
-
-`<home>/.gemini/extensions/my-extension/gemini-extension.json`
-
-### `gemini-extension.json`
-
-The `gemini-extension.json` file contains the configuration for the extension.
-The file has the following structure:
-
-```json
-{
-  "name": "my-extension",
-  "version": "1.0.0",
-  "mcpServers": {
-    "my-server": {
-      "command": "node my-server.js"
-    }
-  },
-  "contextFileName": "GEMINI.md",
-  "excludeTools": ["run_shell_command"]
-}
-```
-
-- `name`: The name of the extension. This is used to uniquely identify the
-  extension and for conflict resolution when extension commands have the same
-  name as user or project commands. The name should be lowercase or numbers and
-  use dashes instead of underscores or spaces. This is how users will refer to
-  your extension in the CLI. Note that we expect this name to match the
-  extension directory name.
-- `version`: The version of the extension.
-- `mcpServers`: A map of MCP servers to settings. The key is the name of the
-  server, and the value is the server configuration. These servers will be
-  loaded on startup just like MCP servers settings in a
-  [`settings.json` file](../get-started/configuration.md). If both an extension
-  and a `settings.json` file settings an MCP server with the same name, the
-  server defined in the `settings.json` file takes precedence.
-  - Note that all MCP server configuration options are supported except for
-    `trust`.
-- `contextFileName`: The name of the file that contains the context for the
-  extension. This will be used to load the context from the extension directory.
-  If this property is not used but a `GEMINI.md` file is present in your
-  extension directory, then that file will be loaded.
-- `excludeTools`: An array of tool names to exclude from the model. You can also
-  specify command-specific restrictions for tools that support it, like the
-  `run_shell_command` tool. For example,
-  `"excludeTools": ["run_shell_command(rm -rf)"]` will block the `rm -rf`
-  command. Note that this differs from the MCP server `excludeTools`
-  functionality, which can be listed in the MCP server config.
-
-When Gemini CLI starts, it loads all the extensions and merges their
-configurations. If there are any conflicts, the workspace configuration takes
-precedence.
-
-### Settings
-
-_Note: This is an experimental feature. We do not yet recommend extension
-authors introduce settings as part of their core flows._
-
-Extensions can define settings that the user will be prompted to provide upon
-installation. This is useful for things like API keys, URLs, or other
-configuration that the extension needs to function.
-
-To define settings, add a `settings` array to your `gemini-extension.json` file.
-Each object in the array should have the following properties:
-
-- `name`: A user-friendly name for the setting.
-- `description`: A description of the setting and what it's used for.
-- `envVar`: The name of the environment variable that the setting will be stored
-  as.
-- `sensitive`: Optional boolean. If true, obfuscates the input the user provides
-  and stores the secret in keychain storage. **Example**
-
-```json
-{
-  "name": "my-api-extension",
-  "version": "1.0.0",
-  "settings": [
-    {
-      "name": "API Key",
-      "description": "Your API key for the service.",
-      "envVar": "MY_API_KEY"
-    }
-  ]
-}
-```
-
-When a user installs this extension, they will be prompted to enter their API
-key. The value will be saved to a `.env` file in the extension's directory
-(e.g., `<home>/.gemini/extensions/my-api-extension/.env`).
-
-You can view a list of an extension's settings by running:
-
-```
+```bash
 gemini extensions list
 ```
 
-and you can update a given setting using:
+## Installation
 
-```
-gemini extensions config <extension name> [setting name] [--scope <scope>]
+To install a real extension, you can use the `extensions install` command with a
+GitHub repository URL in noninteractive mode. For example:
+
+```bash
+gemini extensions install https://github.com/gemini-cli-extensions/workspace
 ```
 
-- `--scope`: The scope to set the setting in (`user` or `workspace`). This is
-  optional and will default to `user`.
+## Next steps
 
-### Custom commands
-
-Extensions can provide [custom commands](../cli/custom-commands.md) by placing
-TOML files in a `commands/` subdirectory within the extension directory. These
-commands follow the same format as user and project custom commands and use
-standard naming conventions.
-
-**Example**
-
-An extension named `gcp` with the following structure:
-
-```
-.gemini/extensions/gcp/
-├── gemini-extension.json
-└── commands/
-    ├── deploy.toml
-    └── gcs/
-        └── sync.toml
-```
-
-Would provide these commands:
-
-- `/deploy` - Shows as `[gcp] Custom command from deploy.toml` in help
-- `/gcs:sync` - Shows as `[gcp] Custom command from sync.toml` in help
-
-### Hooks
-
-Extensions can provide [hooks](../hooks/index.md) to intercept and customize
-Gemini CLI behavior at specific lifecycle events. Hooks provided by an extension
-must be defined in a `hooks/hooks.json` file within the extension directory.
-
-> [!IMPORTANT] Hooks are not defined directly in `gemini-extension.json`. The
-> CLI specifically looks for the `hooks/hooks.json` file.
-
-#### Directory structure
-
-```
-.gemini/extensions/my-extension/
-├── gemini-extension.json
-└── hooks/
-    └── hooks.json
-```
-
-#### `hooks/hooks.json` format
-
-The `hooks.json` file contains a `hooks` object where keys are
-[event names](../hooks/reference.md#supported-events) and values are arrays of
-[hook definitions](../hooks/reference.md#hook-definition).
-
-```json
-{
-  "hooks": {
-    "before_agent": [
-      {
-        "hooks": [
-          {
-            "type": "command",
-            "command": "node ${extensionPath}/scripts/setup.js",
-            "name": "Extension Setup"
-          }
-        ]
-      }
-    ]
-  }
-}
-```
-
-#### Supported variables
-
-Just like `gemini-extension.json`, the `hooks/hooks.json` file supports
-[variable substitution](#variables). This is particularly useful for referencing
-scripts within the extension directory using `${extensionPath}`.
-
-### Conflict resolution
-
-Extension commands have the lowest precedence. When a conflict occurs with user
-or project commands:
-
-1. **No conflict**: Extension command uses its natural name (e.g., `/deploy`)
-2. **With conflict**: Extension command is renamed with the extension prefix
-   (e.g., `/gcp.deploy`)
-
-For example, if both a user and the `gcp` extension define a `deploy` command:
-
-- `/deploy` - Executes the user's deploy command
-- `/gcp.deploy` - Executes the extension's deploy command (marked with `[gcp]`
-  tag)
-
-### Variables
-
-Gemini CLI extensions allow variable substitution in both
-`gemini-extension.json` and `hooks/hooks.json`. This can be useful if e.g., you
-need the current directory to run an MCP server or hook script using
-`"cwd": "${extensionPath}${/}run.ts"`.
-
-**Supported variables:**
-
-| variable                   | description                                                                                                                                                     |
-| -------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------- |
-| `${extensionPath}`         | The fully-qualified path of the extension in the user's filesystem e.g., '/Users/username/.gemini/extensions/example-extension'. This will not unwrap symlinks. |
-| `${workspacePath}`         | The fully-qualified path of the current workspace.                                                                                                              |
-| `${/} or ${pathSeparator}` | The path separator (differs per OS).                                                                                                                            |
-| `${process.execPath}`      | The path to the Node.js binary executing the CLI.                                                                                                               |
+- [Writing extensions](writing-extensions.md): Learn how to create your first
+  extension.
+- [Extensions reference](reference.md): Deeply understand the extension format,
+  commands, and configuration.
+- [Best practices](best-practices.md): Learn strategies for building great
+  extensions.
+- [Extensions releasing](releasing.md): Learn how to share your extensions with
+  the world.

docs/extensions/reference.md

@@ -0,0 +1,312 @@
+# Extensions reference
+
+This guide covers the `gemini extensions` commands and the structure of the
+`gemini-extension.json` configuration file.
+
+## Extension management
+
+We offer a suite of extension management tools using `gemini extensions`
+commands.
+
+Note that these commands (e.g. `gemini extensions install`) are not supported
+from within the CLI's **interactive mode**, although you can list installed
+extensions using the `/extensions list` slash command.
+
+Note that all of these management operations (including updates to slash
+commands) will only be reflected in active CLI sessions on **restart**.
+
+### Installing an extension
+
+You can install an extension using `gemini extensions install` with either a
+GitHub URL or a local path.
+
+Note that we create a copy of the installed extension, so you will need to run
+`gemini extensions update` to pull in changes from both locally-defined
+extensions and those on GitHub.
+
+NOTE: If you are installing an extension from GitHub, you'll need to have `git`
+installed on your machine. See
+[git installation instructions](https://git-scm.com/book/en/v2/Getting-Started-Installing-Git)
+for help.
+
+```
+gemini extensions install <source> [--ref <ref>] [--auto-update] [--pre-release] [--consent]
+```
+
+- `<source>`: The github URL or local path of the extension to install.
+- `--ref`: The git ref to install from.
+- `--auto-update`: Enable auto-update for this extension.
+- `--pre-release`: Enable pre-release versions for this extension.
+- `--consent`: Acknowledge the security risks of installing an extension and
+  skip the confirmation prompt.
+
+### Uninstalling an extension
+
+To uninstall one or more extensions, run
+`gemini extensions uninstall <name...>`:
+
+```
+gemini extensions uninstall gemini-cli-security gemini-cli-another-extension
+```
+
+### Disabling an extension
+
+Extensions are, by default, enabled across all workspaces. You can disable an
+extension entirely or for specific workspace.
+
+```
+gemini extensions disable <name> [--scope <scope>]
+```
+
+- `<name>`: The name of the extension to disable.
+- `--scope`: The scope to disable the extension in (`user` or `workspace`).
+
+### Enabling an extension
+
+You can enable extensions using `gemini extensions enable <name>`. You can also
+enable an extension for a specific workspace using
+`gemini extensions enable <name> --scope=workspace` from within that workspace.
+
+```
+gemini extensions enable <name> [--scope <scope>]
+```
+
+- `<name>`: The name of the extension to enable.
+- `--scope`: The scope to enable the extension in (`user` or `workspace`).
+
+### Updating an extension
+
+For extensions installed from a local path or a git repository, you can
+explicitly update to the latest version (as reflected in the
+`gemini-extension.json` `version` field) with `gemini extensions update <name>`.
+
+You can update all extensions with:
+
+```
+gemini extensions update --all
+```
+
+### Create a boilerplate extension
+
+We offer several example extensions `context`, `custom-commands`,
+`exclude-tools` and `mcp-server`. You can view these examples
+[here](https://github.com/google-gemini/gemini-cli/tree/main/packages/cli/src/commands/extensions/examples).
+
+To copy one of these examples into a development directory using the type of
+your choosing, run:
+
+```
+gemini extensions new <path> [template]
+```
+
+- `<path>`: The path to create the extension in.
+- `[template]`: The boilerplate template to use.
+
+### Link a local extension
+
+The `gemini extensions link` command will create a symbolic link from the
+extension installation directory to the development path.
+
+This is useful so you don't have to run `gemini extensions update` every time
+you make changes you'd like to test.
+
+```
+gemini extensions link <path>
+```
+
+- `<path>`: The path of the extension to link.
+
+## Extension format
+
+On startup, Gemini CLI looks for extensions in `<home>/.gemini/extensions`
+
+Extensions exist as a directory that contains a `gemini-extension.json` file.
+For example:
+
+`<home>/.gemini/extensions/my-extension/gemini-extension.json`
+
+### `gemini-extension.json`
+
+The `gemini-extension.json` file contains the configuration for the extension.
+The file has the following structure:
+
+```json
+{
+  "name": "my-extension",
+  "version": "1.0.0",
+  "description": "My awesome extension",
+  "mcpServers": {
+    "my-server": {
+      "command": "node my-server.js"
+    }
+  },
+  "contextFileName": "GEMINI.md",
+  "excludeTools": ["run_shell_command"]
+}
+```
+
+- `name`: The name of the extension. This is used to uniquely identify the
+  extension and for conflict resolution when extension commands have the same
+  name as user or project commands. The name should be lowercase or numbers and
+  use dashes instead of underscores or spaces. This is how users will refer to
+  your extension in the CLI. Note that we expect this name to match the
+  extension directory name.
+- `version`: The version of the extension.
+- `description`: A short description of the extension. This will be displayed on
+  [geminicli.com/extensions](https://geminicli.com/extensions).
+- `mcpServers`: A map of MCP servers to settings. The key is the name of the
+  server, and the value is the server configuration. These servers will be
+  loaded on startup just like MCP servers settingsd in a
+  [`settings.json` file](../get-started/configuration.md). If both an extension
+  and a `settings.json` file settings an MCP server with the same name, the
+  server defined in the `settings.json` file takes precedence.
+  - Note that all MCP server configuration options are supported except for
+    `trust`.
+- `contextFileName`: The name of the file that contains the context for the
+  extension. This will be used to load the context from the extension directory.
+  If this property is not used but a `GEMINI.md` file is present in your
+  extension directory, then that file will be loaded.
+- `excludeTools`: An array of tool names to exclude from the model. You can also
+  specify command-specific restrictions for tools that support it, like the
+  `run_shell_command` tool. For example,
+  `"excludeTools": ["run_shell_command(rm -rf)"]` will block the `rm -rf`
+  command. Note that this differs from the MCP server `excludeTools`
+  functionality, which can be listed in the MCP server config.
+
+When Gemini CLI starts, it loads all the extensions and merges their
+configurations. If there are any conflicts, the workspace configuration takes
+precedence.
+
+### Settings
+
+_Note: This is an experimental feature. We do not yet recommend extension
+authors introduce settings as part of their core flows._
+
+Extensions can define settings that the user will be prompted to provide upon
+installation. This is useful for things like API keys, URLs, or other
+configuration that the extension needs to function.
+
+To define settings, add a `settings` array to your `gemini-extension.json` file.
+Each object in the array should have the following properties:
+
+- `name`: A user-friendly name for the setting.
+- `description`: A description of the setting and what it's used for.
+- `envVar`: The name of the environment variable that the setting will be stored
+  as.
+- `sensitive`: Optional boolean. If true, obfuscates the input the user provides
+  and stores the secret in keychain storage. **Example**
+
+```json
+{
+  "name": "my-api-extension",
+  "version": "1.0.0",
+  "settings": [
+    {
+      "name": "API Key",
+      "description": "Your API key for the service.",
+      "envVar": "MY_API_KEY"
+    }
+  ]
+}
+```
+
+When a user installs this extension, they will be prompted to enter their API
+key. The value will be saved to a `.env` file in the extension's directory
+(e.g., `<home>/.gemini/extensions/my-api-extension/.env`).
+
+You can view a list of an extension's settings by running:
+
+```
+gemini extensions list
+```
+
+and you can update a given setting using:
+
+```
+gemini extensions config <extension name> [setting name] [--scope <scope>]
+```
+
+- `--scope`: The scope to set the setting in (`user` or `workspace`). This is
+  optional and will default to `user`.
+
+### Custom commands
+
+Extensions can provide [custom commands](../cli/custom-commands.md) by placing
+TOML files in a `commands/` subdirectory within the extension directory. These
+commands follow the same format as user and project custom commands and use
+standard naming conventions.
+
+**Example**
+
+An extension named `gcp` with the following structure:
+
+```
+.gemini/extensions/gcp/
+├── gemini-extension.json
+└── commands/
+    ├── deploy.toml
+    └── gcs/
+        └── sync.toml
+```
+
+Would provide these commands:
+
+- `/deploy` - Shows as `[gcp] Custom command from deploy.toml` in help
+- `/gcs:sync` - Shows as `[gcp] Custom command from sync.toml` in help
+
+### Hooks
+
+Extensions can provide [hooks](../hooks/index.md) to intercept and customize
+Gemini CLI behavior at specific lifecycle events. Hooks provided by an extension
+must be defined in a `hooks/hooks.json` file within the extension directory.
+
+> [!IMPORTANT] Hooks are not defined directly in `gemini-extension.json`. The
+> CLI specifically looks for the `hooks/hooks.json` file.
+
+### Agent Skills
+
+Extensions can bundle [Agent Skills](../cli/skills.md) to provide specialized
+workflows. Skills must be placed in a `skills/` directory within the extension.
+
+**Example**
+
+An extension with the following structure:
+
+```
+.gemini/extensions/my-extension/
+├── gemini-extension.json
+└── skills/
+    └── security-audit/
+        └── SKILL.md
+```
+
+Will expose a `security-audit` skill that the model can activate.
+
+### Conflict resolution
+
+Extension commands have the lowest precedence. When a conflict occurs with user
+or project commands:
+
+1. **No conflict**: Extension command uses its natural name (e.g., `/deploy`)
+2. **With conflict**: Extension command is renamed with the extension prefix
+   (e.g., `/gcp.deploy`)
+
+For example, if both a user and the `gcp` extension define a `deploy` command:
+
+- `/deploy` - Executes the user's deploy command
+- `/gcp.deploy` - Executes the extension's deploy command (marked with `[gcp]`
+  tag)
+
+## Variables
+
+Gemini CLI extensions allow variable substitution in `gemini-extension.json`.
+This can be useful if e.g., you need the current directory to run an MCP server
+using an argument like `"args": ["${extensionPath}${/}dist${/}server.js"]`.
+
+**Supported variables:**
+
+| variable                   | description                                                                                                                                                     |
+| -------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------- |
+| `${extensionPath}`         | The fully-qualified path of the extension in the user's filesystem e.g., '/Users/username/.gemini/extensions/example-extension'. This will not unwrap symlinks. |
+| `${workspacePath}`         | The fully-qualified path of the current workspace.                                                                                                              |
+| `${/} or ${pathSeparator}` | The path separator (differs per OS).                                                                                                                            |

docs/extensions/writing-extensions.md

@@ -8,7 +8,19 @@ file.
 ## Prerequisites
 
 Before you start, make sure you have the Gemini CLI installed and a basic
-understanding of Node.js and TypeScript.
+understanding of Node.js.
+
+## When to use what
+
+Extensions offer a variety of ways to customize Gemini CLI.
+
+| Feature                                                        | What it is                                                                                                         | When to use it                                                                                                                                                                                                                                                                                 | Invoked by            |
+| :------------------------------------------------------------- | :----------------------------------------------------------------------------------------------------------------- | :--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | :-------------------- |
+| **[MCP server](reference.md#mcp-servers)**                     | A standard way to expose new tools and data sources to the model.                                                  | Use this when you want the model to be able to _do_ new things, like fetching data from an internal API, querying a database, or controlling a local application. We also support MCP resources (which can replace custom commands) and system instructions (which can replace custom context) | Model                 |
+| **[Custom commands](../cli/custom-commands.md)**               | A shortcut (like `/my-cmd`) that executes a pre-defined prompt or shell command.                                   | Use this for repetitive tasks or to save long, complex prompts that you use frequently. Great for automation.                                                                                                                                                                                  | User                  |
+| **[Context file (`GEMINI.md`)](reference.md#contextfilename)** | A markdown file containing instructions that are loaded into the model's context at the start of every session.    | Use this to define the "personality" of your extension, set coding standards, or provide essential knowledge that the model should always have.                                                                                                                                                | CLI provides to model |
+| **[Agent skills](../cli/skills.md)**                           | A specialized set of instructions and workflows that the model activates only when needed.                         | Use this for complex, occasional tasks (like "create a PR" or "audit security") to avoid cluttering the main context window when the skill isn't being used.                                                                                                                                   | Model                 |
+| **[Hooks](../hooks/index.md)**                                 | A way to intercept and customize the CLI's behavior at specific lifecycle events (e.g., before/after a tool call). | Use this when you want to automate actions based on what the model is doing, like validating tool arguments, logging activity, or modifying the model's input/output.                                                                                                                          | CLI                   |
 
 ## Step 1: Create a new extension
 
@@ -26,10 +38,9 @@ This will create a new directory with the following structure:
 
 ```
 my-first-extension/
-├── example.ts
+├── example.js
 ├── gemini-extension.json
-├── package.json
-└── tsconfig.json
+└── package.json
 ```
 
 ## Step 2: Understand the extension files
@@ -43,12 +54,12 @@ and use your extension.
 
 ```json
 {
-  "name": "my-first-extension",
+  "name": "mcp-server-example",
   "version": "1.0.0",
   "mcpServers": {
     "nodeServer": {
       "command": "node",
-      "args": ["${extensionPath}${/}dist${/}example.js"],
+      "args": ["${extensionPath}${/}example.js"],
       "cwd": "${extensionPath}"
     }
   }
@@ -64,12 +75,12 @@ and use your extension.
     with the absolute path to your extension's installation directory. This
     allows your extension to work regardless of where it's installed.
 
-### `example.ts`
+### `example.js`
 
 This file contains the source code for your MCP server. It's a simple Node.js
 server that uses the `@modelcontextprotocol/sdk`.
 
-```typescript
+```javascript
 /**
  * @license
  * Copyright 2025 Google LLC
@@ -118,16 +129,15 @@ await server.connect(transport);
 This server defines a single tool called `fetch_posts` that fetches data from a
 public API.
 
-### `package.json` and `tsconfig.json`
+### `package.json`
 
-These are standard configuration files for a TypeScript project. The
-`package.json` file defines dependencies and a `build` script, and
-`tsconfig.json` configures the TypeScript compiler.
+This is the standard configuration file for a Node.js project. It defines
+dependencies and scripts.
 
-## Step 3: Build and link your extension
+## Step 3: Link your extension
 
-Before you can use the extension, you need to compile the TypeScript code and
-link the extension to your Gemini CLI installation for local development.
+Before you can use the extension, you need to link it to your Gemini CLI
+installation for local development.
 
 1.  **Install dependencies:**
 
@@ -136,16 +146,7 @@ link the extension to your Gemini CLI installation for local development.
     npm install
     ```
 
-2.  **Build the server:**
-
-    ```bash
-    npm run build
-    ```
-
-    This will compile `example.ts` into `dist/example.js`, which is the file
-    referenced in your `gemini-extension.json`.
-
-3.  **Link the extension:**
+2.  **Link the extension:**
 
     The `link` command creates a symbolic link from the Gemini CLI extensions
     directory to your development directory. This means any changes you make
@@ -212,7 +213,7 @@ need this for extensions built to expose commands and prompts.
       "mcpServers": {
         "nodeServer": {
           "command": "node",
-          "args": ["${extensionPath}${/}dist${/}example.js"],
+          "args": ["${extensionPath}${/}example.js"],
           "cwd": "${extensionPath}"
         }
       }
@@ -222,14 +223,50 @@ need this for extensions built to expose commands and prompts.
 Restart the CLI again. The model will now have the context from your `GEMINI.md`
 file in every session where the extension is active.
 
-## Step 6: Releasing your extension
+## (Optional) Step 6: Add an Agent Skill
 
-Once you are happy with your extension, you can share it with others. The two
+_Note: This is an experimental feature enabled via `experimental.skills`._
+
+[Agent Skills](../cli/skills.md) let you bundle specialized expertise and
+procedural workflows. Unlike `GEMINI.md`, which provides persistent context,
+skills are activated only when needed, saving context tokens.
+
+1.  Create a `skills` directory and a subdirectory for your skill:
+
+    ```bash
+    mkdir -p skills/security-audit
+    ```
+
+2.  Create a `skills/security-audit/SKILL.md` file:
+
+    ```markdown
+    ---
+    name: security-audit
+    description:
+      Expertise in auditing code for security vulnerabilities. Use when the user
+      asks to "check for security issues" or "audit" their changes.
+    ---
+
+    # Security Auditor
+
+    You are an expert security researcher. When auditing code:
+
+    1. Look for common vulnerabilities (OWASP Top 10).
+    2. Check for hardcoded secrets or API keys.
+    3. Suggest remediation steps for any findings.
+    ```
+
+Skills bundled with your extension are automatically discovered and can be
+activated by the model during a session when it identifies a relevant task.
+
+## Step 7: Release your extension
+
+Once you're happy with your extension, you can share it with others. The two
 primary ways of releasing extensions are via a Git repository or through GitHub
 Releases. Using a public Git repository is the simplest method.
 
 For detailed instructions on both methods, please refer to the
-[Extension Releasing Guide](./extension-releasing.md).
+[Extension Releasing Guide](./releasing.md).
 
 ## Conclusion
 
@@ -239,6 +276,7 @@ You've successfully created a Gemini CLI extension! You learned how to:
 - Add custom tools with an MCP server.
 - Create convenient custom commands.
 - Provide persistent context to the model.
+- Bundle specialized Agent Skills.
 - Link your extension for local development.
 
 From here, you can explore more advanced features and build powerful new

docs/get-started/configuration.md

@@ -110,13 +110,13 @@ their corresponding top-level category object in your `settings.json` file.
   - **Description:** Enable Vim keybindings
   - **Default:** `false`
 
-- **`general.disableAutoUpdate`** (boolean):
-  - **Description:** Disable automatic updates
-  - **Default:** `false`
+- **`general.enableAutoUpdate`** (boolean):
+  - **Description:** Enable automatic updates.
+  - **Default:** `true`
 
-- **`general.disableUpdateNag`** (boolean):
-  - **Description:** Disable update notification prompts.
-  - **Default:** `false`
+- **`general.enableAutoUpdateNotification`** (boolean):
+  - **Description:** Enable update notification prompts.
+  - **Default:** `true`
 
 - **`general.checkpointing.enabled`** (boolean):
   - **Description:** Enable session checkpointing for recovery
@@ -159,7 +159,7 @@ their corresponding top-level category object in your `settings.json` file.
 #### `output`
 
 - **`output.format`** (enum):
-  - **Description:** The format of the CLI output.
+  - **Description:** The format of the CLI output. Can be `text` or `json`.
   - **Default:** `"text"`
   - **Values:** `"text"`, `"json"`
 
@@ -180,10 +180,21 @@ their corresponding top-level category object in your `settings.json` file.
   - **Requires restart:** Yes
 
 - **`ui.showStatusInTitle`** (boolean):
-  - **Description:** Show Gemini CLI status and thoughts in the terminal window
-    title
+  - **Description:** Show Gemini CLI model thoughts in the terminal window title
+    during the working phase
   - **Default:** `false`
 
+- **`ui.dynamicWindowTitle`** (boolean):
+  - **Description:** Update the terminal window title with current status icons
+    (Ready: ◇, Action Required: ✋, Working: ✦)
+  - **Default:** `true`
+
+- **`ui.showHomeDirectoryWarning`** (boolean):
+  - **Description:** Show a warning when running Gemini CLI in the home
+    directory.
+  - **Default:** `true`
+  - **Requires restart:** Yes
+
 - **`ui.hideTips`** (boolean):
   - **Description:** Hide helpful tips in the UI
   - **Default:** `false`
@@ -233,19 +244,15 @@ their corresponding top-level category object in your `settings.json` file.
   - **Description:** Show the model name in the chat for each model turn.
   - **Default:** `false`
 
-- **`ui.useFullWidth`** (boolean):
-  - **Description:** Use the entire width of the terminal for output.
-  - **Default:** `true`
-
 - **`ui.useAlternateBuffer`** (boolean):
   - **Description:** Use an alternate screen buffer for the UI, preserving shell
     history.
   - **Default:** `false`
   - **Requires restart:** Yes
 
-- **`ui.compact`** (boolean):
-  - **Description:** Enable a more compact UI layout.
-  - **Default:** `false`
+- **`ui.useBackgroundColor`** (boolean):
+  - **Description:** Whether to use background colors in the UI.
+  - **Default:** `true`
 
 - **`ui.incrementalRendering`** (boolean):
   - **Description:** Enable incremental rendering for the UI. This option will
@@ -254,14 +261,18 @@ their corresponding top-level category object in your `settings.json` file.
   - **Default:** `true`
   - **Requires restart:** Yes
 
+- **`ui.showSpinner`** (boolean):
+  - **Description:** Show the spinner during operations.
+  - **Default:** `true`
+
 - **`ui.customWittyPhrases`** (array):
   - **Description:** Custom witty phrases to display during loading. When
     provided, the CLI cycles through these instead of the defaults.
   - **Default:** `[]`
 
-- **`ui.accessibility.disableLoadingPhrases`** (boolean):
-  - **Description:** Disable loading phrases for accessibility
-  - **Default:** `false`
+- **`ui.accessibility.enableLoadingPhrases`** (boolean):
+  - **Description:** Enable loading phrases during operations.
+  - **Default:** `true`
   - **Requires restart:** Yes
 
 - **`ui.accessibility.screenReader`** (boolean):
@@ -273,7 +284,7 @@ their corresponding top-level category object in your `settings.json` file.
 #### `ide`
 
 - **`ide.enabled`** (boolean):
-  - **Description:** Enable IDE integration mode
+  - **Description:** Enable IDE integration mode.
   - **Default:** `false`
   - **Requires restart:** Yes
 
@@ -550,6 +561,14 @@ their corresponding top-level category object in your `settings.json` file.
     used.
   - **Default:** `[]`
 
+#### `agents`
+
+- **`agents.overrides`** (object):
+  - **Description:** Override settings for specific agents, e.g. to disable the
+    agent, set a custom model config, or run config.
+  - **Default:** `{}`
+  - **Requires restart:** Yes
+
 #### `context`
 
 - **`context.fileName`** (string | string[]):
@@ -577,12 +596,12 @@ their corresponding top-level category object in your `settings.json` file.
   - **Default:** `false`
 
 - **`context.fileFiltering.respectGitIgnore`** (boolean):
-  - **Description:** Respect .gitignore files when searching
+  - **Description:** Respect .gitignore files when searching.
   - **Default:** `true`
   - **Requires restart:** Yes
 
 - **`context.fileFiltering.respectGeminiIgnore`** (boolean):
-  - **Description:** Respect .geminiignore files when searching
+  - **Description:** Respect .geminiignore files when searching.
   - **Default:** `true`
   - **Requires restart:** Yes
 
@@ -592,9 +611,9 @@ their corresponding top-level category object in your `settings.json` file.
   - **Default:** `true`
   - **Requires restart:** Yes
 
-- **`context.fileFiltering.disableFuzzySearch`** (boolean):
-  - **Description:** Disable fuzzy search when searching for files.
-  - **Default:** `false`
+- **`context.fileFiltering.enableFuzzySearch`** (boolean):
+  - **Description:** Enable fuzzy search when searching for files.
+  - **Default:** `true`
   - **Requires restart:** Yes
 
 #### `tools`
@@ -635,6 +654,13 @@ their corresponding top-level category object in your `settings.json` file.
     considered safe (e.g., read-only operations).
   - **Default:** `false`
 
+- **`tools.approvalMode`** (enum):
+  - **Description:** The default approval mode for tool execution. 'default'
+    prompts for approval, 'auto_edit' auto-approves edit tools, and 'plan' is
+    read-only mode. 'yolo' is not supported yet.
+  - **Default:** `"default"`
+  - **Values:** `"default"`, `"auto_edit"`, `"plan"`
+
 - **`tools.core`** (array):
   - **Description:** Restrict the set of built-in tools with an allowlist. Match
     semantics mirror tools.allowed; see the built-in tools documentation for
@@ -688,6 +714,13 @@ their corresponding top-level category object in your `settings.json` file.
   - **Default:** `1000`
   - **Requires restart:** Yes
 
+- **`tools.disableLLMCorrection`** (boolean):
+  - **Description:** Disable LLM-based error correction for edit tools. When
+    enabled, tools will fail immediately if exact string matches are not found,
+    instead of attempting to self-correct.
+  - **Default:** `true`
+  - **Requires restart:** Yes
+
 - **`tools.enableHooks`** (boolean):
   - **Description:** Enables the hooks system experiment. When disabled, the
     hooks system is completely deactivated regardless of other settings.
@@ -808,6 +841,16 @@ their corresponding top-level category object in your `settings.json` file.
   - **Default:** `true`
   - **Requires restart:** Yes
 
+- **`experimental.extensionConfig`** (boolean):
+  - **Description:** Enable requesting and fetching of extension settings.
+  - **Default:** `false`
+  - **Requires restart:** Yes
+
+- **`experimental.enableEventDrivenScheduler`** (boolean):
+  - **Description:** Enables event-driven scheduler within the CLI session.
+  - **Default:** `true`
+  - **Requires restart:** Yes
+
 - **`experimental.extensionReloading`** (boolean):
   - **Description:** Enables extension loading/unloading within the CLI session.
   - **Default:** `false`
@@ -823,41 +866,14 @@ their corresponding top-level category object in your `settings.json` file.
   - **Default:** `false`
   - **Requires restart:** Yes
 
-- **`experimental.codebaseInvestigatorSettings.enabled`** (boolean):
-  - **Description:** Enable the Codebase Investigator agent.
-  - **Default:** `true`
-  - **Requires restart:** Yes
-
-- **`experimental.codebaseInvestigatorSettings.maxNumTurns`** (number):
-  - **Description:** Maximum number of turns for the Codebase Investigator
-    agent.
-  - **Default:** `10`
-  - **Requires restart:** Yes
-
-- **`experimental.codebaseInvestigatorSettings.maxTimeMinutes`** (number):
-  - **Description:** Maximum time for the Codebase Investigator agent (in
-    minutes).
-  - **Default:** `3`
-  - **Requires restart:** Yes
-
-- **`experimental.codebaseInvestigatorSettings.thinkingBudget`** (number):
-  - **Description:** The thinking budget for the Codebase Investigator agent.
-  - **Default:** `8192`
-  - **Requires restart:** Yes
-
-- **`experimental.codebaseInvestigatorSettings.model`** (string):
-  - **Description:** The model to use for the Codebase Investigator agent.
-  - **Default:** `"auto"`
-  - **Requires restart:** Yes
-
 - **`experimental.useOSC52Paste`** (boolean):
   - **Description:** Use OSC 52 sequence for pasting instead of clipboardy
     (useful for remote sessions).
   - **Default:** `false`
 
-- **`experimental.cliHelpAgentSettings.enabled`** (boolean):
-  - **Description:** Enable the CLI Help Agent.
-  - **Default:** `true`
+- **`experimental.plan`** (boolean):
+  - **Description:** Enable planning features (Plan Mode and tools).
+  - **Default:** `false`
   - **Requires restart:** Yes
 
 #### `skills`
@@ -867,22 +883,24 @@ their corresponding top-level category object in your `settings.json` file.
   - **Default:** `[]`
   - **Requires restart:** Yes
 
-#### `hooks`
+#### `hooksConfig`
 
-- **`hooks.enabled`** (boolean):
+- **`hooksConfig.enabled`** (boolean):
   - **Description:** Canonical toggle for the hooks system. When disabled, no
     hooks will be executed.
-  - **Default:** `false`
+  - **Default:** `true`
 
-- **`hooks.disabled`** (array):
+- **`hooksConfig.disabled`** (array):
   - **Description:** List of hook names (commands) that should be disabled.
     Hooks in this list will not execute even if configured.
   - **Default:** `[]`
 
-- **`hooks.notifications`** (boolean):
+- **`hooksConfig.notifications`** (boolean):
   - **Description:** Show visual indicators when hooks are executing.
   - **Default:** `true`
 
+#### `hooks`
+
 - **`hooks.BeforeTool`** (array):
   - **Description:** Hooks that execute before tool execution. Can intercept,
     validate, or modify tool calls.
@@ -952,6 +970,10 @@ their corresponding top-level category object in your `settings.json` file.
 - **`admin.mcp.enabled`** (boolean):
   - **Description:** If false, disallows MCP servers from being used.
   - **Default:** `true`
+
+- **`admin.skills.enabled`** (boolean):
+  - **Description:** If false, disallows agent skills from being used.
+  - **Default:** `true`
   <!-- SETTINGS-AUTOGEN:END -->
 
 #### `mcpServers`
@@ -1129,7 +1151,7 @@ the `advanced.excludedEnvVars` setting in your `settings.json` file.
 - **`GEMINI_MODEL`**:
   - Specifies the default Gemini model to use.
   - Overrides the hardcoded default
-  - Example: `export GEMINI_MODEL="gemini-2.5-flash"`
+  - Example: `export GEMINI_MODEL="gemini-3-flash-preview"`
 - **`GOOGLE_API_KEY`**:
   - Your Google Cloud API key.
   - Required for using Vertex AI in express mode.
@@ -1274,7 +1296,7 @@ for that specific session.
 
 - **`--model <model_name>`** (**`-m <model_name>`**):
   - Specifies the Gemini model to use for this session.
-  - Example: `npm start -- --model gemini-1.5-pro-latest`
+  - Example: `npm start -- --model gemini-3-pro-preview`
 - **`--prompt <your_prompt>`** (**`-p <your_prompt>`**):
   - Used to pass a prompt directly to the command. This invokes Gemini CLI in a
     non-interactive mode.
@@ -1297,7 +1319,8 @@ for that specific session.
 - **`--sandbox`** (**`-s`**):
   - Enables sandbox mode for this session.
 - **`--debug`** (**`-d`**):
-  - Enables debug mode for this session, providing more verbose output.
+  - Enables debug mode for this session, providing more verbose output. Open the
+    debug console with F12 to see the additional logging.
 
 - **`--help`** (or **`-h`**):
   - Displays help information about command-line arguments.
@@ -1309,6 +1332,10 @@ for that specific session.
     - `auto_edit`: Automatically approve edit tools (replace, write_file) while
       prompting for others
     - `yolo`: Automatically approve all tool calls (equivalent to `--yolo`)
+    - `plan`: Read-only mode for tool calls (requires experimental planning to
+      be enabled).
+      > **Note:** This mode is currently under development and not yet fully
+      > functional.
   - Cannot be used together with `--yolo`. Use `--approval-mode=yolo` instead of
     `--yolo` for the new unified approach.
   - Example: `gemini --approval-mode auto_edit`

docs/hooks/best-practices.md

@@ -1,4 +1,4 @@
-# Hooks on Gemini CLI: Best practices
+# Hooks Best Practices
 
 This guide covers security considerations, performance optimization, debugging
 techniques, and privacy considerations for developing and deploying hooks in
@@ -15,21 +15,20 @@ using parallel operations:
 // Sequential operations are slower
 const data1 = await fetch(url1).then((r) => r.json());
 const data2 = await fetch(url2).then((r) => r.json());
-const data3 = await fetch(url3).then((r) => r.json());
 
 // Prefer parallel operations for better performance
 // Start requests concurrently
 const p1 = fetch(url1).then((r) => r.json());
 const p2 = fetch(url2).then((r) => r.json());
-const p3 = fetch(url3).then((r) => r.json());
 
 // Wait for all results
-const [data1, data2, data3] = await Promise.all([p1, p2, p3]);
+const [data1, data2] = await Promise.all([p1, p2]);
 ```
 
 ### Cache expensive operations
 
-Store results between invocations to avoid repeated computation:
+Store results between invocations to avoid repeated computation, especially for
+hooks that run frequently (like `BeforeTool` or `AfterModel`).
 
 ```javascript
 const fs = require('fs');
@@ -54,6 +53,7 @@ async function main() {
   const cacheKey = `tool-list-${(Date.now() / 3600000) | 0}`; // Hourly cache
 
   if (cache[cacheKey]) {
+    // Write JSON to stdout
     console.log(JSON.stringify(cache[cacheKey]));
     return;
   }
@@ -70,32 +70,20 @@ async function main() {
 ### Use appropriate events
 
 Choose hook events that match your use case to avoid unnecessary execution.
-`AfterAgent` fires once per agent loop completion, while `AfterModel` fires
-after every LLM call (potentially multiple times per loop):
 
-```json
-// If checking final completion, use AfterAgent instead of AfterModel
-{
-  "hooks": {
-    "AfterAgent": [
-      {
-        "matcher": "*",
-        "hooks": [
-          {
-            "name": "final-checker",
-            "command": "./check-completion.sh"
-          }
-        ]
-      }
-    ]
-  }
-}
-```
+- **`AfterAgent`**: Fires **once** per turn after the model finishes its final
+  response. Use this for quality validation (Retries) or final logging.
+- **`AfterModel`**: Fires after **every chunk** of LLM output. Use this for
+  real-time redaction, PII filtering, or monitoring output as it streams.
+
+If you only need to check the final completion, use `AfterAgent` to save
+performance.
 
 ### Filter with matchers
 
 Use specific matchers to avoid unnecessary hook execution. Instead of matching
-all tools with `*`, specify only the tools you need:
+all tools with `*`, specify only the tools you need. This saves the overhead of
+spawning a process for irrelevant events.
 
 ```json
 {
@@ -103,6 +91,7 @@ all tools with `*`, specify only the tools you need:
   "hooks": [
     {
       "name": "validate-writes",
+      "type": "command",
       "command": "./validate.sh"
     }
   ]
@@ -111,30 +100,32 @@ all tools with `*`, specify only the tools you need:
 
 ### Optimize JSON parsing
 
-For large inputs, use streaming JSON parsers to avoid loading everything into
-memory:
-
-```javascript
-// Standard approach: parse entire input
-const input = JSON.parse(await readStdin());
-const content = input.tool_input.content;
-
-// For very large inputs: stream and extract only needed fields
-const { createReadStream } = require('fs');
-const JSONStream = require('JSONStream');
-
-const stream = createReadStream(0).pipe(JSONStream.parse('tool_input.content'));
-let content = '';
-stream.on('data', (chunk) => {
-  content += chunk;
-});
-```
+For large inputs (like `AfterModel` receiving a large context), standard JSON
+parsing can be slow. If you only need one field, consider streaming parsers or
+lightweight extraction logic, though for most shell scripts `jq` is sufficient.
 
 ## Debugging
 
+### The "Strict JSON" rule
+
+The most common cause of hook failure is "polluting" the standard output.
+
+- **stdout** is for **JSON only**.
+- **stderr** is for **logs and text**.
+
+**Good:**
+
+```bash
+#!/bin/bash
+echo "Starting check..." >&2  # <--- Redirect to stderr
+echo '{"decision": "allow"}'
+
+```
+
 ### Log to files
 
-Write debug information to dedicated log files:
+Since hooks run in the background, writing to a dedicated log file is often the
+easiest way to debug complex logic.
 
 ```bash
 #!/usr/bin/env bash
@@ -151,6 +142,9 @@ log "Received input: ${input:0:100}..."
 # Hook logic here
 
 log "Hook completed successfully"
+# Always output valid JSON to stdout at the end, even if just empty
+echo "{}"
+
 ```
 
 ### Use stderr for errors
@@ -162,6 +156,7 @@ try {
   const result = dangerousOperation();
   console.log(JSON.stringify({ result }));
 } catch (error) {
+  // Write the error description to stderr so the user/agent sees it
   console.error(`Hook error: ${error.message}`);
   process.exit(2); // Blocking error
 }
@@ -169,7 +164,8 @@ try {
 
 ### Test hooks independently
 
-Run hook scripts manually with sample JSON input:
+Run hook scripts manually with sample JSON input to verify they behave as
+expected before hooking them up to the CLI.
 
 ```bash
 # Create test input
@@ -191,33 +187,46 @@ cat test-input.json | .gemini/hooks/my-hook.sh
 
 # Check exit code
 echo "Exit code: $?"
+
 ```
 
 ### Check exit codes
 
-Ensure your script returns the correct exit code:
+Gemini CLI uses exit codes for high-level flow control:
+
+- **Exit 0 (Success)**: The hook ran successfully. The CLI parses `stdout` for
+  JSON decisions.
+- **Exit 2 (System Block)**: A critical block occurred. `stderr` is used as the
+  reason.
+  - For **Agent/Model** events, this aborts the turn.
+  - For **Tool** events, this blocks the tool but allows the agent to continue.
+  - For **AfterAgent**, this triggers an automatic retry turn.
+
+> **TIP**
+>
+> **Blocking vs. Stopping**: Use `decision: "deny"` (or Exit Code 2) to block a
+> **specific action**. Use `{"continue": false}` in your JSON output to **kill
+> the entire agent loop** immediately.
 
 ```bash
 #!/usr/bin/env bash
-set -e  # Exit on error
+set -e
 
 # Hook logic
-process_input() {
-  # ...
-}
-
 if process_input; then
-  echo "Success message"
+  echo '{"decision": "allow"}'
   exit 0
 else
-  echo "Error message" >&2
+  echo "Critical validation failure" >&2
   exit 2
 fi
+
 ```
 
 ### Enable telemetry
 
-Hook execution is logged when `telemetry.logPrompts` is enabled:
+Hook execution is logged when `telemetry.logPrompts` is enabled. You can view
+these logs to debug execution flow.
 
 ```json
 {
@@ -227,11 +236,10 @@ Hook execution is logged when `telemetry.logPrompts` is enabled:
 }
 ```
 
-View hook telemetry in logs to debug execution issues.
-
 ### Use hook panel
 
-The `/hooks panel` command shows execution status and recent output:
+The `/hooks panel` command inside the CLI shows execution status and recent
+output:
 
 ```bash
 /hooks panel
@@ -255,18 +263,64 @@ Begin with basic logging hooks before implementing complex logic:
 # Simple logging hook to understand input structure
 input=$(cat)
 echo "$input" >> .gemini/hook-inputs.log
-echo "Logged input"
+# Always return valid JSON
+echo "{}"
+
+```
+
+### Documenting your hooks
+
+Maintainability is critical for complex hook systems. Use descriptions and
+comments to help yourself and others understand why a hook exists.
+
+**Use the `description` field**: This text is displayed in the `/hooks panel` UI
+and helps diagnose issues.
+
+```json
+{
+  "hooks": {
+    "BeforeTool": [
+      {
+        "matcher": "write_file|replace",
+        "hooks": [
+          {
+            "name": "secret-scanner",
+            "type": "command",
+            "command": "$GEMINI_PROJECT_DIR/.gemini/hooks/block-secrets.sh",
+            "description": "Scans code changes for API keys and secrets before writing"
+          }
+        ]
+      }
+    ]
+  }
+}
+```
+
+**Add comments in hook scripts**: Explain performance expectations and
+dependencies.
+
+```javascript
+#!/usr/bin/env node
+/**
+ * RAG Tool Filter Hook
+ *
+ * Reduces the tool space by extracting keywords from the user's request.
+ *
+ * Performance: ~500ms average
+ * Dependencies: @google/generative-ai
+ */
 ```
 
 ### Use JSON libraries
 
-Parse JSON with proper libraries instead of text processing:
+Parse JSON with proper libraries instead of text processing.
 
 **Bad:**
 
 ```bash
 # Fragile text parsing
 tool_name=$(echo "$input" | grep -oP '"tool_name":\s*"\K[^"]+')
+
 ```
 
 **Good:**
@@ -274,6 +328,7 @@ tool_name=$(echo "$input" | grep -oP '"tool_name":\s*"\K[^"]+')
 ```bash
 # Robust JSON parsing
 tool_name=$(echo "$input" | jq -r '.tool_name')
+
 ```
 
 ### Make scripts executable
@@ -283,6 +338,7 @@ Always make hook scripts executable:
 ```bash
 chmod +x .gemini/hooks/*.sh
 chmod +x .gemini/hooks/*.js
+
 ```
 
 ### Version control
@@ -292,7 +348,7 @@ Commit hooks to share with your team:
 ```bash
 git add .gemini/hooks/
 git add .gemini/settings.json
-git commit -m "Add project hooks for security and testing"
+
 ```
 
 **`.gitignore` considerations:**
@@ -306,295 +362,10 @@ git commit -m "Add project hooks for security and testing"
 # Keep hook scripts
 !.gemini/hooks/*.sh
 !.gemini/hooks/*.js
+
 ```
 
-### Document behavior
-
-Add descriptions to help others understand your hooks:
-
-```json
-{
-  "hooks": {
-    "BeforeTool": [
-      {
-        "matcher": "write_file|replace",
-        "hooks": [
-          {
-            "name": "secret-scanner",
-            "type": "command",
-            "command": "$GEMINI_PROJECT_DIR/.gemini/hooks/block-secrets.sh",
-            "description": "Scans code changes for API keys, passwords, and other secrets before writing"
-          }
-        ]
-      }
-    ]
-  }
-}
-```
-
-Add comments in hook scripts:
-
-```javascript
-#!/usr/bin/env node
-/**
- * RAG Tool Filter Hook
- *
- * This hook reduces the tool space from 100+ tools to ~15 relevant ones
- * by extracting keywords from the user's request and filtering tools
- * based on semantic similarity.
- *
- * Performance: ~500ms average, cached tool embeddings
- * Dependencies: @google/generative-ai
- */
-```
-
-## Troubleshooting
-
-### Hook not executing
-
-**Check hook name in `/hooks panel`:**
-
-```bash
-/hooks panel
-```
-
-Verify the hook appears in the list and is enabled.
-
-**Verify matcher pattern:**
-
-```bash
-# Test regex pattern
-echo "write_file|replace" | grep -E "write_.*|replace"
-```
-
-**Check disabled list:**
-
-```json
-{
-  "hooks": {
-    "disabled": ["my-hook-name"]
-  }
-}
-```
-
-**Ensure script is executable:**
-
-```bash
-ls -la .gemini/hooks/my-hook.sh
-chmod +x .gemini/hooks/my-hook.sh
-```
-
-**Verify script path:**
-
-```bash
-# Check path expansion
-echo "$GEMINI_PROJECT_DIR/.gemini/hooks/my-hook.sh"
-
-# Verify file exists
-test -f "$GEMINI_PROJECT_DIR/.gemini/hooks/my-hook.sh" && echo "File exists"
-```
-
-### Hook timing out
-
-**Check configured timeout:**
-
-```json
-{
-  "name": "slow-hook",
-  "timeout": 60000
-}
-```
-
-**Optimize slow operations:**
-
-```javascript
-// Before: Sequential operations (slow)
-for (const item of items) {
-  await processItem(item);
-}
-
-// After: Parallel operations (fast)
-await Promise.all(items.map((item) => processItem(item)));
-```
-
-**Use caching:**
-
-```javascript
-const cache = new Map();
-
-async function getCachedData(key) {
-  if (cache.has(key)) {
-    return cache.get(key);
-  }
-  const data = await fetchData(key);
-  cache.set(key, data);
-  return data;
-}
-```
-
-**Consider splitting into multiple faster hooks:**
-
-```json
-{
-  "hooks": {
-    "BeforeTool": [
-      {
-        "matcher": "write_file",
-        "hooks": [
-          {
-            "name": "quick-check",
-            "command": "./quick-validation.sh",
-            "timeout": 1000
-          }
-        ]
-      },
-      {
-        "matcher": "write_file",
-        "hooks": [
-          {
-            "name": "deep-check",
-            "command": "./deep-analysis.sh",
-            "timeout": 30000
-          }
-        ]
-      }
-    ]
-  }
-}
-```
-
-### Invalid JSON output
-
-**Validate JSON before outputting:**
-
-```bash
-#!/usr/bin/env bash
-output='{"decision": "allow"}'
-
-# Validate JSON
-if echo "$output" | jq empty 2>/dev/null; then
-  echo "$output"
-else
-  echo "Invalid JSON generated" >&2
-  exit 1
-fi
-```
-
-**Ensure proper quoting and escaping:**
-
-```javascript
-// Bad: Unescaped string interpolation
-const message = `User said: ${userInput}`;
-console.log(JSON.stringify({ message }));
-
-// Good: Automatic escaping
-console.log(JSON.stringify({ message: `User said: ${userInput}` }));
-```
-
-**Check for binary data or control characters:**
-
-```javascript
-function sanitizeForJSON(str) {
-  return str.replace(/[\x00-\x1F\x7F-\x9F]/g, ''); // Remove control chars
-}
-
-const cleanContent = sanitizeForJSON(content);
-console.log(JSON.stringify({ content: cleanContent }));
-```
-
-### Exit code issues
-
-**Verify script returns correct codes:**
-
-```bash
-#!/usr/bin/env bash
-set -e  # Exit on error
-
-# Processing logic
-if validate_input; then
-  echo "Success"
-  exit 0
-else
-  echo "Validation failed" >&2
-  exit 2
-fi
-```
-
-**Check for unintended errors:**
-
-```bash
-#!/usr/bin/env bash
-# Don't use 'set -e' if you want to handle errors explicitly
-# set -e
-
-if ! command_that_might_fail; then
-  # Handle error
-  echo "Command failed but continuing" >&2
-fi
-
-# Always exit explicitly
-exit 0
-```
-
-**Use trap for cleanup:**
-
-```bash
-#!/usr/bin/env bash
-
-cleanup() {
-  # Cleanup logic
-  rm -f /tmp/hook-temp-*
-}
-
-trap cleanup EXIT
-
-# Hook logic here
-```
-
-### Environment variables not available
-
-**Check if variable is set:**
-
-```bash
-#!/usr/bin/env bash
-
-if [ -z "$GEMINI_PROJECT_DIR" ]; then
-  echo "GEMINI_PROJECT_DIR not set" >&2
-  exit 1
-fi
-
-if [ -z "$CUSTOM_VAR" ]; then
-  echo "Warning: CUSTOM_VAR not set, using default" >&2
-  CUSTOM_VAR="default-value"
-fi
-```
-
-**Debug available variables:**
-
-```bash
-#!/usr/bin/env bash
-
-# List all environment variables
-env > .gemini/hook-env.log
-
-# Check specific variables
-echo "GEMINI_PROJECT_DIR: $GEMINI_PROJECT_DIR" >> .gemini/hook-env.log
-echo "GEMINI_SESSION_ID: $GEMINI_SESSION_ID" >> .gemini/hook-env.log
-echo "GEMINI_API_KEY: ${GEMINI_API_KEY:+<set>}" >> .gemini/hook-env.log
-```
-
-**Use .env files:**
-
-```bash
-#!/usr/bin/env bash
-
-# Load .env file if it exists
-if [ -f "$GEMINI_PROJECT_DIR/.env" ]; then
-  source "$GEMINI_PROJECT_DIR/.env"
-fi
-```
-
-## Using Hooks Securely
+## Hook security
 
 ### Threat Model
 
@@ -621,11 +392,10 @@ When you open a project with hooks defined in `.gemini/settings.json`:
    it).
 5. **Trust**: The hook is marked as "trusted" for this project.
 
-> [!IMPORTANT] **Modification Detection**: If the `command` string of a project
-> hook is changed (e.g., by a `git pull`), its identity changes. Gemini CLI will
-> treat it as a **new, untrusted hook** and warn you again. This prevents
-> malicious actors from silently swapping a verified command for a malicious
-> one.
+> **Modification detection**: If the `command` string of a project hook is
+> changed (e.g., by a `git pull`), its identity changes. Gemini CLI will treat
+> it as a **new, untrusted hook** and warn you again. This prevents malicious
+> actors from silently swapping a verified command for a malicious one.
 
 ### Risks
 
@@ -646,32 +416,134 @@ When you open a project with hooks defined in `.gemini/settings.json`:
   publishers, well-known community members).
 - Be cautious with obfuscated scripts or compiled binaries from unknown sources.
 
-#### Sanitize Environment
+#### Sanitize environment
 
 Hooks inherit the environment of the Gemini CLI process, which may include
-sensitive API keys. Gemini CLI attempts to sanitize sensitive variables, but you
-should be cautious.
+sensitive API keys. Gemini CLI provides a
+[redaction system](/docs/get-started/configuration#environment-variable-redaction)
+that automatically filters variables matching sensitive patterns (e.g., `KEY`,
+`TOKEN`).
 
-- **Avoid printing environment variables** to stdout/stderr unless necessary.
-- **Use `.env` files** to securely manage sensitive variables, ensuring they are
-  excluded from version control.
+> **Disabled by Default**: Environment redaction is currently **OFF by
+> default**. We strongly recommend enabling it if you are running third-party
+> hooks or working in sensitive environments.
 
-**System Administrators:** You can enforce environment variable redaction by
-default in the system configuration (e.g., `/etc/gemini-cli/settings.json`):
+**Impact on hooks:**
+
+- **Security**: Prevents your hook scripts from accidentally leaking secrets.
+- **Troubleshooting**: If your hook depends on a specific environment variable
+  that is being blocked, you must explicitly allow it in `settings.json`.
 
 ```json
 {
   "security": {
     "environmentVariableRedaction": {
       "enabled": true,
-      "blocked": ["MY_SECRET_KEY"],
-      "allowed": ["SAFE_VAR"]
+      "allowed": ["MY_REQUIRED_TOOL_KEY"]
     }
   }
 }
 ```
 
-## Authoring Secure Hooks
+**System administrators:** You can enforce redaction for all users in the system
+configuration.
+
+## Troubleshooting
+
+### Hook not executing
+
+**Check hook name in `/hooks panel`:** Verify the hook appears in the list and
+is enabled.
+
+**Verify matcher pattern:**
+
+```bash
+# Test regex pattern
+echo "write_file|replace" | grep -E "write_.*|replace"
+
+```
+
+**Check disabled list:** Verify the hook is not listed in your `settings.json`:
+
+```json
+{
+  "hooks": {
+    "disabled": ["my-hook-name"]
+  }
+}
+```
+
+**Ensure script is executable**: For macOS and Linux users, verify the script
+has execution permissions:
+
+```bash
+ls -la .gemini/hooks/my-hook.sh
+chmod +x .gemini/hooks/my-hook.sh
+```
+
+**Verify script path:** Ensure the path in `settings.json` resolves correctly.
+
+```bash
+# Check path expansion
+echo "$GEMINI_PROJECT_DIR/.gemini/hooks/my-hook.sh"
+
+# Verify file exists
+test -f "$GEMINI_PROJECT_DIR/.gemini/hooks/my-hook.sh" && echo "File exists"
+```
+
+### Hook timing out
+
+**Check configured timeout:** The default is 60000ms (1 minute). You can
+increase this in `settings.json`:
+
+```json
+{
+  "name": "slow-hook",
+  "timeout": 120000
+}
+```
+
+**Optimize slow operations:** Move heavy processing to background tasks or use
+caching.
+
+### Invalid JSON output
+
+**Validate JSON before outputting:**
+
+```bash
+#!/usr/bin/env bash
+output='{"decision": "allow"}'
+
+# Validate JSON
+if echo "$output" | jq empty 2>/dev/null; then
+  echo "$output"
+else
+  echo "Invalid JSON generated" >&2
+  exit 1
+fi
+
+```
+
+### Environment variables not available
+
+**Check if variable is set:**
+
+```bash
+#!/usr/bin/env bash
+if [ -z "$GEMINI_PROJECT_DIR" ]; then
+  echo "GEMINI_PROJECT_DIR not set" >&2
+  exit 1
+fi
+
+```
+
+**Debug available variables:**
+
+```bash
+env > .gemini/hook-env.log
+```
+
+## Authoring secure hooks
 
 When writing your own hooks, follow these practices to ensure they are robust
 and secure.
@@ -713,6 +585,7 @@ defaults to 60 seconds, but you should set stricter limits for fast hooks.
         "hooks": [
           {
             "name": "fast-validator",
+            "type": "command",
             "command": "./hooks/validate.sh",
             "timeout": 5000 // 5 seconds
           }
@@ -766,40 +639,17 @@ function containsSecret(content) {
 
 ## Privacy considerations
 
-Hook inputs and outputs may contain sensitive information. Gemini CLI respects
-the `telemetry.logPrompts` setting for hook data logging.
+Hook inputs and outputs may contain sensitive information.
 
 ### What data is collected
 
-Hook telemetry may include:
-
-- **Hook inputs:** User prompts, tool arguments, file contents
-- **Hook outputs:** Hook responses, decision reasons, added context
-- **Standard streams:** stdout and stderr from hook processes
-- **Execution metadata:** Hook name, event type, duration, success/failure
+Hook telemetry may include inputs (prompts, code) and outputs (decisions,
+reasons) unless disabled.
 
 ### Privacy settings
 
-**Enabled (default):**
-
-Full hook I/O is logged to telemetry. Use this when:
-
-- Developing and debugging hooks
-- Telemetry is redirected to a trusted enterprise system
-- You understand and accept the privacy implications
-
-**Disabled:**
-
-Only metadata is logged (event name, duration, success/failure). Hook inputs and
-outputs are excluded. Use this when:
-
-- Sending telemetry to third-party systems
-- Working with sensitive data
-- Privacy regulations require minimizing data collection
-
-### Configuration
-
-**Disable PII logging in settings:**
+**Disable PII logging:** If you are working with sensitive data, disable prompt
+logging in your settings:
 
 ```json
 {
@@ -809,48 +659,19 @@ outputs are excluded. Use this when:
 }
 ```
 
-**Disable via environment variable:**
+**Suppress Output:** Individual hooks can request their metadata be hidden from
+logs and telemetry by returning `"suppressOutput": true` in their JSON response.
 
-```bash
-export GEMINI_TELEMETRY_LOG_PROMPTS=false
-```
+> **Note**
+
+> `suppressOutput` only affects background logging. Any `systemMessage` or
+> `reason` included in the JSON will still be displayed to the user in the
+> terminal.
 
 ### Sensitive data in hooks
 
 If your hooks process sensitive data:
 
-1. **Minimize logging:** Don't write sensitive data to log files
-2. **Sanitize outputs:** Remove sensitive data before outputting
-3. **Use secure storage:** Encrypt sensitive data at rest
-4. **Limit access:** Restrict hook script permissions
-
-**Example sanitization:**
-
-```javascript
-function sanitizeOutput(data) {
-  const sanitized = { ...data };
-
-  // Remove sensitive fields
-  delete sanitized.apiKey;
-  delete sanitized.password;
-
-  // Redact sensitive strings
-  if (sanitized.content) {
-    sanitized.content = sanitized.content.replace(
-      /api[_-]?key\s*[:=]\s*['"]?[a-zA-Z0-9_-]{20,}['"]?/gi,
-      '[REDACTED]',
-    );
-  }
-
-  return sanitized;
-}
-
-console.log(JSON.stringify(sanitizeOutput(hookOutput)));
-```
-
-## Learn more
-
-- [Hooks Reference](index.md) - Complete API reference
-- [Writing Hooks](writing-hooks.md) - Tutorial and examples
-- [Configuration](../get-started/configuration.md) - Gemini CLI settings
-- [Hooks Design Document](../hooks-design.md) - Technical architecture
+1. **Minimize logging:** Don't write sensitive data to log files.
+2. **Sanitize outputs:** Remove sensitive data before outputting JSON or writing
+   to stderr.

docs/hooks/index.md

@@ -1,99 +1,118 @@
-# Gemini CLI hooks
+# Gemini CLI hooks (experimental)
 
 Hooks are scripts or programs that Gemini CLI executes at specific points in the
 agentic loop, allowing you to intercept and customize behavior without modifying
 the CLI's source code.
 
-See [writing hooks guide](writing-hooks.md) for a tutorial on creating your
-first hook and a comprehensive example.
+## Availability
 
-See [hooks reference](reference.md) for the technical specification of the I/O
-schemas.
+> **Experimental Feature**: Hooks are currently enabled by default only in the
+> **Preview** and **Nightly** release channels.
 
-See [best practices](best-practices.md) for guidelines on security, performance,
-and debugging.
+If you are on the Stable channel, you must explicitly enable the hooks system in
+your `settings.json`:
+
+```json
+{
+  "hooksConfig": {
+    "enabled": true
+  }
+}
+```
+
+- **[Writing hooks guide](/docs/hooks/writing-hooks)**: A tutorial on creating
+  your first hook with comprehensive examples.
+- **[Hooks reference](/docs/hooks/reference)**: The definitive technical
+  specification of I/O schemas and exit codes.
+- **[Best practices](/docs/hooks/best-practices)**: Guidelines on security,
+  performance, and debugging.
 
 ## What are hooks?
 
-With hooks, you can:
-
-- **Add context:** Inject relevant information before the model processes a
-  request
-- **Validate actions:** Review and block potentially dangerous operations
-- **Enforce policies:** Implement security and compliance requirements
-- **Log interactions:** Track tool usage and model responses
-- **Optimize behavior:** Dynamically adjust tool selection or model parameters
-
 Hooks run synchronously as part of the agent loop—when a hook event fires,
 Gemini CLI waits for all matching hooks to complete before continuing.
 
-## Security and Risks
+With hooks, you can:
 
-> [!WARNING] **Hooks execute arbitrary code with your user privileges.**
-
-By configuring hooks, you are explicitly allowing Gemini CLI to run shell
-commands on your machine. Malicious or poorly configured hooks can:
-
-- **Exfiltrate data**: Read sensitive files (`.env`, ssh keys) and send them to
-  remote servers.
-- **Modify system**: Delete files, install malware, or change system settings.
-- **Consume resources**: Run infinite loops or crash your system.
-
-**Project-level hooks** (in `.gemini/settings.json`) and **Extension hooks** are
-particularly risky when opening third-party projects or extensions from
-untrusted authors. Gemini CLI will **warn you** the first time it detects a new
-project hook (identified by its name and command), but it is **your
-responsibility** to review these hooks (and any installed extensions) before
-trusting them.
-
-> [!NOTE] Extension hooks are subject to a mandatory security warning and
-> consent flow during extension installation or update if hooks are detected.
-> You must explicitly approve the installation or update of any extension that
-> contains hooks.
-
-See [Security Considerations](best-practices.md#using-hooks-securely) for a
-detailed threat model and mitigation strategies.
+- **Add context:** Inject relevant information (like git history) before the
+  model processes a request.
+- **Validate actions:** Review tool arguments and block potentially dangerous
+  operations.
+- **Enforce policies:** Implement security scanners and compliance checks.
+- **Log interactions:** Track tool usage and model responses for auditing.
+- **Optimize behavior:** Dynamically filter available tools or adjust model
+  parameters.
 
 ## Core concepts
 
 ### Hook events
 
-Hooks are triggered by specific events in Gemini CLI's lifecycle. The following
-table lists all available hook events:
+Hooks are triggered by specific events in Gemini CLI's lifecycle.
 
-| Event                 | When It Fires                                 | Common Use Cases                           |
-| --------------------- | --------------------------------------------- | ------------------------------------------ |
-| `SessionStart`        | When a session begins                         | Initialize resources, load context         |
-| `SessionEnd`          | When a session ends                           | Clean up, save state                       |
-| `BeforeAgent`         | After user submits prompt, before planning    | Add context, validate prompts              |
-| `AfterAgent`          | When agent loop ends                          | Review output, force continuation          |
-| `BeforeModel`         | Before sending request to LLM                 | Modify prompts, add instructions           |
-| `AfterModel`          | After receiving LLM response                  | Filter responses, log interactions         |
-| `BeforeToolSelection` | Before LLM selects tools (after BeforeModel)  | Filter available tools, optimize selection |
-| `BeforeTool`          | Before a tool executes                        | Validate arguments, block dangerous ops    |
-| `AfterTool`           | After a tool executes                         | Process results, run tests                 |
-| `PreCompress`         | Before context compression                    | Save state, notify user                    |
-| `Notification`        | When a notification occurs (e.g., permission) | Auto-approve, log decisions                |
+| Event                 | When It Fires                                  | Impact                 | Common Use Cases                             |
+| --------------------- | ---------------------------------------------- | ---------------------- | -------------------------------------------- |
+| `SessionStart`        | When a session begins (startup, resume, clear) | Inject Context         | Initialize resources, load context           |
+| `SessionEnd`          | When a session ends (exit, clear)              | Advisory               | Clean up, save state                         |
+| `BeforeAgent`         | After user submits prompt, before planning     | Block Turn / Context   | Add context, validate prompts, block turns   |
+| `AfterAgent`          | When agent loop ends                           | Retry / Halt           | Review output, force retry or halt execution |
+| `BeforeModel`         | Before sending request to LLM                  | Block Turn / Mock      | Modify prompts, swap models, mock responses  |
+| `AfterModel`          | After receiving LLM response                   | Block Turn / Redact    | Filter/redact responses, log interactions    |
+| `BeforeToolSelection` | Before LLM selects tools                       | Filter Tools           | Filter available tools, optimize selection   |
+| `BeforeTool`          | Before a tool executes                         | Block Tool / Rewrite   | Validate arguments, block dangerous ops      |
+| `AfterTool`           | After a tool executes                          | Block Result / Context | Process results, run tests, hide results     |
+| `PreCompress`         | Before context compression                     | Advisory               | Save state, notify user                      |
+| `Notification`        | When a system notification occurs              | Advisory               | Forward to desktop alerts, logging           |
 
-### Hook types
+### Global mechanics
 
-Gemini CLI currently supports **command hooks** that run shell commands or
-scripts:
+Understanding these core principles is essential for building robust hooks.
 
-```json
-{
-  "type": "command",
-  "command": "$GEMINI_PROJECT_DIR/.gemini/hooks/my-hook.sh",
-  "timeout": 30000
-}
-```
+#### Strict JSON requirements (The "Golden Rule")
 
-**Note:** Plugin hooks (npm packages) are planned for a future release.
+Hooks communicate via `stdin` (Input) and `stdout` (Output).
 
-### Matchers
+1. **Silence is Mandatory**: Your script **must not** print any plain text to
+   `stdout` other than the final JSON object. **Even a single `echo` or `print`
+   call before the JSON will break parsing.**
+2. **Pollution = Failure**: If `stdout` contains non-JSON text, parsing will
+   fail. The CLI will default to "Allow" and treat the entire output as a
+   `systemMessage`.
+3. **Debug via Stderr**: Use `stderr` for **all** logging and debugging (e.g.,
+   `echo "debug" >&2`). Gemini CLI captures `stderr` but never attempts to parse
+   it as JSON.
 
-For tool-related events (`BeforeTool`, `AfterTool`), you can filter which tools
-trigger the hook:
+#### Exit codes
+
+Gemini CLI uses exit codes to determine the high-level outcome of a hook
+execution:
+
+| Exit Code | Label            | Behavioral Impact                                                                                                                                                            |
+| --------- | ---------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
+| **0**     | **Success**      | The `stdout` is parsed as JSON. **Preferred code** for all logic, including intentional blocks (e.g., `{"decision": "deny"}`).                                               |
+| **2**     | **System Block** | **Critical Block**. The target action (tool, turn, or stop) is aborted. `stderr` is used as the rejection reason. High severity; used for security stops or script failures. |
+| **Other** | **Warning**      | Non-fatal failure. A warning is shown, but the interaction proceeds using original parameters.                                                                               |
+
+#### Matchers
+
+You can filter which specific tools or triggers fire your hook using the
+`matcher` field.
+
+- **Tool events** (`BeforeTool`, `AfterTool`): Matchers are **Regular
+  Expressions**. (e.g., `"write_.*"`).
+- **Lifecycle events**: Matchers are **Exact Strings**. (e.g., `"startup"`).
+- **Wildcards**: `"*"` or `""` (empty string) matches all occurrences.
+
+## Configuration
+
+Hooks are configured in `settings.json`. Gemini CLI merges configurations from
+multiple layers in the following order of precedence (highest to lowest):
+
+1.  **Project settings**: `.gemini/settings.json` in the current directory.
+2.  **User settings**: `~/.gemini/settings.json`.
+3.  **System settings**: `/etc/gemini-cli/settings.json`.
+4.  **Extensions**: Hooks defined by installed extensions.
+
+### Configuration schema
 
 ```json
 {
@@ -101,382 +120,12 @@ trigger the hook:
     "BeforeTool": [
       {
         "matcher": "write_file|replace",
-        "hooks": [
-          /* hooks for write operations */
-        ]
-      }
-    ]
-  }
-}
-```
-
-**Matcher patterns:**
-
-- **Exact match:** `"read_file"` matches only `read_file`
-- **Regex:** `"write_.*|replace"` matches `write_file`, `replace`
-- **Wildcard:** `"*"` or `""` matches all tools
-
-**Session event matchers:**
-
-- **SessionStart:** `startup`, `resume`, `clear`
-- **SessionEnd:** `exit`, `clear`, `logout`, `prompt_input_exit`
-- **PreCompress:** `manual`, `auto`
-- **Notification:** `ToolPermission`
-
-## Hook input/output contract
-
-### Command hook communication
-
-Hooks communicate via:
-
-- **Input:** JSON on stdin
-- **Output:** Exit code + stdout/stderr
-
-### Exit codes
-
-- **0:** Success - stdout shown to user (or injected as context for some events)
-- **2:** Blocking error - stderr shown to agent/user, operation may be blocked
-- **Other:** Non-blocking warning - logged but execution continues
-
-### Common input fields
-
-Every hook receives these base fields:
-
-```json
-{
-  "session_id": "abc123",
-  "transcript_path": "/path/to/transcript.jsonl",
-  "cwd": "/path/to/project",
-  "hook_event_name": "BeforeTool",
-  "timestamp": "2025-12-01T10:30:00Z"
-  // ... event-specific fields
-}
-```
-
-### Event-specific fields
-
-#### BeforeTool
-
-**Input:**
-
-```json
-{
-  "tool_name": "write_file",
-  "tool_input": {
-    "file_path": "/path/to/file.ts",
-    "content": "..."
-  }
-}
-```
-
-**Output (JSON on stdout):**
-
-```json
-{
-  "decision": "allow|deny|ask|block",
-  "reason": "Explanation shown to agent",
-  "systemMessage": "Message shown to user"
-}
-```
-
-Or simple exit codes:
-
-- Exit 0 = allow (stdout shown to user)
-- Exit 2 = deny (stderr shown to agent)
-
-#### AfterTool
-
-**Input:**
-
-```json
-{
-  "tool_name": "read_file",
-  "tool_input": { "file_path": "..." },
-  "tool_response": "file contents..."
-}
-```
-
-**Output:**
-
-```json
-{
-  "decision": "allow|deny",
-  "hookSpecificOutput": {
-    "hookEventName": "AfterTool",
-    "additionalContext": "Extra context for agent"
-  }
-}
-```
-
-#### BeforeAgent
-
-**Input:**
-
-```json
-{
-  "prompt": "Fix the authentication bug"
-}
-```
-
-**Output:**
-
-```json
-{
-  "decision": "allow|deny",
-  "hookSpecificOutput": {
-    "hookEventName": "BeforeAgent",
-    "additionalContext": "Recent project decisions: ..."
-  }
-}
-```
-
-#### BeforeModel
-
-**Input:**
-
-```json
-{
-  "llm_request": {
-    "model": "gemini-2.0-flash-exp",
-    "messages": [{ "role": "user", "content": "Hello" }],
-    "config": { "temperature": 0.7 },
-    "toolConfig": {
-      "functionCallingConfig": {
-        "mode": "AUTO",
-        "allowedFunctionNames": ["read_file", "write_file"]
-      }
-    }
-  }
-}
-```
-
-**Output:**
-
-```json
-{
-  "decision": "allow",
-  "hookSpecificOutput": {
-    "hookEventName": "BeforeModel",
-    "llm_request": {
-      "messages": [
-        { "role": "system", "content": "Additional instructions..." },
-        { "role": "user", "content": "Hello" }
-      ]
-    }
-  }
-}
-```
-
-#### AfterModel
-
-**Input:**
-
-```json
-{
-  "llm_request": {
-    "model": "gemini-2.0-flash-exp",
-    "messages": [
-      /* ... */
-    ],
-    "config": {
-      /* ... */
-    },
-    "toolConfig": {
-      /* ... */
-    }
-  },
-  "llm_response": {
-    "text": "string",
-    "candidates": [
-      {
-        "content": {
-          "role": "model",
-          "parts": ["array of content parts"]
-        },
-        "finishReason": "STOP"
-      }
-    ]
-  }
-}
-```
-
-**Output:**
-
-```json
-{
-  "hookSpecificOutput": {
-    "hookEventName": "AfterModel",
-    "llm_response": {
-      "candidate": {
-        /* modified response */
-      }
-    }
-  }
-}
-```
-
-#### BeforeToolSelection
-
-**Input:**
-
-```json
-{
-  "llm_request": {
-    "model": "gemini-2.0-flash-exp",
-    "messages": [
-      /* ... */
-    ],
-    "toolConfig": {
-      "functionCallingConfig": {
-        "mode": "AUTO",
-        "allowedFunctionNames": [
-          /* 100+ tools */
-        ]
-      }
-    }
-  }
-}
-```
-
-**Output:**
-
-```json
-{
-  "hookSpecificOutput": {
-    "hookEventName": "BeforeToolSelection",
-    "toolConfig": {
-      "functionCallingConfig": {
-        "mode": "ANY",
-        "allowedFunctionNames": ["read_file", "write_file", "replace"]
-      }
-    }
-  }
-}
-```
-
-Or simple output (comma-separated tool names sets mode to ANY):
-
-```bash
-echo "read_file,write_file,replace"
-```
-
-#### SessionStart
-
-**Input:**
-
-```json
-{
-  "source": "startup|resume|clear"
-}
-```
-
-**Output:**
-
-```json
-{
-  "hookSpecificOutput": {
-    "hookEventName": "SessionStart",
-    "additionalContext": "Loaded 5 project memories"
-  }
-}
-```
-
-#### SessionEnd
-
-**Input:**
-
-```json
-{
-  "reason": "exit|clear|logout|prompt_input_exit|other"
-}
-```
-
-No structured output expected (but stdout/stderr logged).
-
-#### PreCompress
-
-**Input:**
-
-```json
-{
-  "trigger": "manual|auto"
-}
-```
-
-**Output:**
-
-```json
-{
-  "systemMessage": "Compression starting..."
-}
-```
-
-#### Notification
-
-**Input:**
-
-```json
-{
-  "notification_type": "ToolPermission",
-  "message": "string",
-  "details": {
-    /* notification details */
-  }
-}
-```
-
-**Output:**
-
-```json
-{
-  "systemMessage": "Notification logged"
-}
-```
-
-## Configuration
-
-Hook definitions are configured in `settings.json` files using the `hooks`
-object. Configuration can be specified at multiple levels with defined
-precedence rules.
-
-### Configuration layers
-
-Hook configurations are applied in the following order of execution (lower
-numbers run first):
-
-1.  **Project settings:** `.gemini/settings.json` in your project directory
-    (highest priority)
-2.  **User settings:** `~/.gemini/settings.json`
-3.  **System settings:** `/etc/gemini-cli/settings.json`
-4.  **Extensions:** Internal hooks defined by installed extensions (lowest
-    priority). See [Extensions documentation](../extensions/index.md#hooks) for
-    details on how extensions define and configure hooks.
-
-#### Deduplication and shadowing
-
-If multiple hooks with the identical **name** and **command** are discovered
-across different configuration layers, Gemini CLI deduplicates them. The hook
-from the higher-priority layer (e.g., Project) will be kept, and others will be
-ignored.
-
-Within each level, hooks run in the order they are declared in the
-configuration.
-
-### Configuration schema
-
-```json
-{
-  "hooks": {
-    "EventName": [
-      {
-        "matcher": "pattern",
         "hooks": [
           {
-            "name": "hook-identifier",
+            "name": "security-check",
             "type": "command",
-            "command": "./path/to/script.sh",
-            "description": "What this hook does",
-            "timeout": 30000
+            "command": "$GEMINI_PROJECT_DIR/.gemini/hooks/security.sh",
+            "timeout": 5000
           }
         ]
       }
@@ -485,210 +134,45 @@ configuration.
 }
 ```
 
-**Configuration properties:**
+#### Hook configuration fields
 
-- **`name`** (string, recommended): Unique identifier for the hook used in
-  `/hooks enable/disable` commands. If omitted, the `command` path is used as
-  the identifier.
-- **`type`** (string, required): Hook type - currently only `"command"` is
-  supported
-- **`command`** (string, required): Path to the script or command to execute
-- **`description`** (string, optional): Human-readable description shown in
-  `/hooks panel`
-- **`timeout`** (number, optional): Timeout in milliseconds (default: 60000)
-- **`matcher`** (string, optional): Pattern to filter when hook runs (event
-  matchers only)
+| Field         | Type   | Required  | Description                                                          |
+| :------------ | :----- | :-------- | :------------------------------------------------------------------- |
+| `type`        | string | **Yes**   | The execution engine. Currently only `"command"` is supported.       |
+| `command`     | string | **Yes\*** | The shell command to execute. (Required when `type` is `"command"`). |
+| `name`        | string | No        | A friendly name for identifying the hook in logs and CLI commands.   |
+| `timeout`     | number | No        | Execution timeout in milliseconds (default: 60000).                  |
+| `description` | string | No        | A brief explanation of the hook's purpose.                           |
+
+---
 
 ### Environment variables
 
-Hooks have access to:
+Hooks are executed with a sanitized environment.
 
-- `GEMINI_PROJECT_DIR`: Project root directory
-- `GEMINI_SESSION_ID`: Current session ID
-- `GEMINI_API_KEY`: Gemini API key (if configured)
-- All other environment variables from the parent process
+- `GEMINI_PROJECT_DIR`: The absolute path to the project root.
+- `GEMINI_SESSION_ID`: The unique ID for the current session.
+- `GEMINI_CWD`: The current working directory.
+- `CLAUDE_PROJECT_DIR`: (Alias) Provided for compatibility.
+
+## Security and risks
+
+> **Warning: Hooks execute arbitrary code with your user privileges.** By
+> configuring hooks, you are allowing scripts to run shell commands on your
+> machine.
+
+**Project-level hooks** are particularly risky when opening untrusted projects.
+Gemini CLI **fingerprints** project hooks. If a hook's name or command changes
+(e.g., via `git pull`), it is treated as a **new, untrusted hook** and you will
+be warned before it executes.
+
+See [Security Considerations](/docs/hooks/best-practices#using-hooks-securely)
+for a detailed threat model.
 
 ## Managing hooks
 
-### View registered hooks
+Use the CLI commands to manage hooks without editing JSON manually:
 
-Use the `/hooks panel` command to view all registered hooks:
-
-```bash
-/hooks panel
-```
-
-This command displays:
-
-- All active hooks organized by event
-- Hook source (user, project, system)
-- Hook type (command or plugin)
-- Execution status and recent output
-
-### Enable and disable hooks
-
-You can temporarily enable or disable individual hooks using commands:
-
-```bash
-/hooks enable hook-name
-/hooks disable hook-name
-```
-
-These commands allow you to control hook execution without editing configuration
-files. The hook name should match the `name` field in your hook configuration.
-Changes made via these commands are persisted to your global User settings
-(`~/.gemini/settings.json`).
-
-### Disabled hooks configuration
-
-To permanently disable hooks, add them to the `hooks.disabled` array in your
-`settings.json`:
-
-```json
-{
-  "hooks": {
-    "disabled": ["secret-scanner", "auto-test"]
-  }
-}
-```
-
-**Note:** The `hooks.disabled` array uses a UNION merge strategy. Disabled hooks
-from all configuration levels (user, project, system) are combined and
-deduplicated, meaning a hook disabled at any level remains disabled.
-
-## Migration from Claude Code
-
-If you have hooks configured for Claude Code, you can migrate them:
-
-```bash
-gemini hooks migrate --from-claude
-```
-
-This command:
-
-- Reads `.claude/settings.json`
-- Converts event names (`PreToolUse` → `BeforeTool`, etc.)
-- Translates tool names (`Bash` → `run_shell_command`, `replace` → `replace`)
-- Updates matcher patterns
-- Writes to `.gemini/settings.json`
-
-### Event name mapping
-
-| Claude Code        | Gemini CLI     |
-| ------------------ | -------------- |
-| `PreToolUse`       | `BeforeTool`   |
-| `PostToolUse`      | `AfterTool`    |
-| `UserPromptSubmit` | `BeforeAgent`  |
-| `Stop`             | `AfterAgent`   |
-| `Notification`     | `Notification` |
-| `SessionStart`     | `SessionStart` |
-| `SessionEnd`       | `SessionEnd`   |
-| `PreCompact`       | `PreCompress`  |
-
-### Tool name mapping
-
-| Claude Code | Gemini CLI            |
-| ----------- | --------------------- |
-| `Bash`      | `run_shell_command`   |
-| `Edit`      | `replace`             |
-| `Read`      | `read_file`           |
-| `Write`     | `write_file`          |
-| `Glob`      | `glob`                |
-| `Grep`      | `search_file_content` |
-| `LS`        | `list_directory`      |
-
-## Tool and Event Matchers Reference
-
-### Available tool names for matchers
-
-The following built-in tools can be used in `BeforeTool` and `AfterTool` hook
-matchers:
-
-#### File operations
-
-- `read_file` - Read a single file
-- `read_many_files` - Read multiple files at once
-- `write_file` - Create or overwrite a file
-- `replace` - Edit file content with find/replace
-
-#### File system
-
-- `list_directory` - List directory contents
-- `glob` - Find files matching a pattern
-- `search_file_content` - Search within file contents
-
-#### Execution
-
-- `run_shell_command` - Execute shell commands
-
-#### Web and external
-
-- `google_web_search` - Google Search with grounding
-- `web_fetch` - Fetch web page content
-
-#### Agent features
-
-- `write_todos` - Manage TODO items
-- `save_memory` - Save information to memory
-- `delegate_to_agent` - Delegate tasks to sub-agents
-
-#### Example matchers
-
-```json
-{
-  "matcher": "write_file|replace" // File editing tools
-}
-```
-
-```json
-{
-  "matcher": "read_.*" // All read operations
-}
-```
-
-```json
-{
-  "matcher": "run_shell_command" // Only shell commands
-}
-```
-
-```json
-{
-  "matcher": "*" // All tools
-}
-```
-
-### Event-specific matchers
-
-#### SessionStart event matchers
-
-- `startup` - Fresh session start
-- `resume` - Resuming a previous session
-- `clear` - Session cleared
-
-#### SessionEnd event matchers
-
-- `exit` - Normal exit
-- `clear` - Session cleared
-- `logout` - User logged out
-- `prompt_input_exit` - Exit from prompt input
-- `other` - Other reasons
-
-#### PreCompress event matchers
-
-- `manual` - Manually triggered compression
-- `auto` - Automatically triggered compression
-
-#### Notification event matchers
-
-- `ToolPermission` - Tool permission notifications
-
-## Learn more
-
-- [Writing Hooks](writing-hooks.md) - Tutorial and comprehensive example
-- [Best Practices](best-practices.md) - Security, performance, and debugging
-- [Custom Commands](../cli/custom-commands.md) - Create reusable prompt
-  shortcuts
-- [Configuration](../get-started/configuration.md) - Gemini CLI configuration
-  options
-- [Hooks Design Document](../hooks-design.md) - Technical architecture details
+- **View hooks:** `/hooks panel`
+- **Enable/Disable all:** `/hooks enable-all` or `/hooks disable-all`
+- **Toggle individual:** `/hooks enable <name>` or `/hooks disable <name>`

docs/hooks/reference.md

@@ -1,178 +1,322 @@
-# Hooks Reference
+# Hooks reference
 
 This document provides the technical specification for Gemini CLI hooks,
-including the JSON schemas for input and output, exit code behaviors, and the
-stable model API.
+including JSON schemas and API details.
 
-## Communication Protocol
+## Global hook mechanics
 
-Hooks communicate with Gemini CLI via standard streams and exit codes:
-
-- **Input**: Gemini CLI sends a JSON object to the hook's `stdin`.
-- **Output**: The hook sends a JSON object (or plain text) to `stdout`.
-- **Exit Codes**: Used to signal success or blocking errors.
-
-### Exit Code Behavior
-
-| Exit Code | Meaning            | Behavior                                                                                        |
-| :-------- | :----------------- | :---------------------------------------------------------------------------------------------- |
-| `0`       | **Success**        | `stdout` is parsed as JSON. If parsing fails, it's treated as a `systemMessage`.                |
-| `2`       | **Blocking Error** | Interrupts the current operation. `stderr` is shown to the agent (for tool events) or the user. |
-| Other     | **Warning**        | Execution continues. `stderr` is logged as a non-blocking warning.                              |
+- **Communication**: `stdin` for Input (JSON), `stdout` for Output (JSON), and
+  `stderr` for logs and feedback.
+- **Exit codes**:
+  - `0`: Success. `stdout` is parsed as JSON. **Preferred for all logic.**
+  - `2`: System Block. The action is blocked; `stderr` is used as the rejection
+    reason.
+  - `Other`: Warning. A non-fatal failure occurred; the CLI continues with a
+    warning.
+- **Silence is Mandatory**: Your script **must not** print any plain text to
+  `stdout` other than the final JSON.
 
 ---
 
-## Input Schema (`stdin`)
+## Configuration schema
 
-Every hook receives a base JSON object. Extra fields are added depending on the
-specific event.
+Hooks are defined in `settings.json` within the `hooks` object. Each event
+(e.g., `BeforeTool`) contains an array of **hook definitions**.
 
-### Base Fields (All Events)
+### Hook definition
 
-| Field             | Type     | Description                                           |
-| :---------------- | :------- | :---------------------------------------------------- |
-| `session_id`      | `string` | Unique identifier for the current CLI session.        |
-| `transcript_path` | `string` | Path to the session's JSON transcript (if available). |
-| `cwd`             | `string` | The current working directory.                        |
-| `hook_event_name` | `string` | The name of the firing event (e.g., `BeforeTool`).    |
-| `timestamp`       | `string` | ISO 8601 timestamp of the event.                      |
+| Field        | Type      | Required | Description                                                                             |
+| :----------- | :-------- | :------- | :-------------------------------------------------------------------------------------- |
+| `matcher`    | `string`  | No       | A regex (for tools) or exact string (for lifecycle) to filter when the hook runs.       |
+| `sequential` | `boolean` | No       | If `true`, hooks in this group run one after another. If `false`, they run in parallel. |
+| `hooks`      | `array`   | **Yes**  | An array of **hook configurations**.                                                    |
 
-### Event-Specific Fields
+### Hook configuration
 
-#### Tool Events (`BeforeTool`, `AfterTool`)
-
-- `tool_name`: (`string`) The internal name of the tool (e.g., `write_file`,
-  `run_shell_command`).
-- `tool_input`: (`object`) The arguments passed to the tool.
-- `tool_response`: (`object`, **AfterTool only**) The raw output from the tool
-  execution.
-- `mcp_context`: (`object`, **optional**) Present only for MCP tool invocations.
-  Contains server identity information:
-  - `server_name`: (`string`) The configured name of the MCP server.
-  - `tool_name`: (`string`) The original tool name from the MCP server.
-  - `command`: (`string`, optional) For stdio transport, the command used to
-    start the server.
-  - `args`: (`string[]`, optional) For stdio transport, the command arguments.
-  - `cwd`: (`string`, optional) For stdio transport, the working directory.
-  - `url`: (`string`, optional) For SSE/HTTP transport, the server URL.
-  - `tcp`: (`string`, optional) For WebSocket transport, the TCP address.
-
-#### Agent Events (`BeforeAgent`, `AfterAgent`)
-
-- `prompt`: (`string`) The user's submitted prompt.
-- `prompt_response`: (`string`, **AfterAgent only**) The final response text
-  from the model.
-- `stop_hook_active`: (`boolean`, **AfterAgent only**) Indicates if a stop hook
-  is already handling a continuation.
-
-#### Model Events (`BeforeModel`, `AfterModel`, `BeforeToolSelection`)
-
-- `llm_request`: (`LLMRequest`) A stable representation of the outgoing request.
-  See [Stable Model API](#stable-model-api).
-- `llm_response`: (`LLMResponse`, **AfterModel only**) A stable representation
-  of the incoming response.
-
-#### Session & Notification Events
-
-- `source`: (`startup` | `resume` | `clear`, **SessionStart only**) The trigger
-  source.
-- `reason`: (`exit` | `clear` | `logout` | `prompt_input_exit` | `other`,
-  **SessionEnd only**) The reason for session end.
-- `trigger`: (`manual` | `auto`, **PreCompress only**) What triggered the
-  compression event.
-- `notification_type`: (`ToolPermission`, **Notification only**) The type of
-  notification being fired.
-- `message`: (`string`, **Notification only**) The notification message.
-- `details`: (`object`, **Notification only**) Payload-specific details for the
-  notification.
+| Field         | Type     | Required  | Description                                                          |
+| :------------ | :------- | :-------- | :------------------------------------------------------------------- |
+| `type`        | `string` | **Yes**   | The execution engine. Currently only `"command"` is supported.       |
+| `command`     | `string` | **Yes\*** | The shell command to execute. (Required when `type` is `"command"`). |
+| `name`        | `string` | No        | A friendly name for identifying the hook in logs and CLI commands.   |
+| `timeout`     | `number` | No        | Execution timeout in milliseconds (default: 60000).                  |
+| `description` | `string` | No        | A brief explanation of the hook's purpose.                           |
 
 ---
 
-## Output Schema (`stdout`)
+## Base input schema
 
-If the hook exits with `0`, the CLI attempts to parse `stdout` as JSON.
+All hooks receive these common fields via `stdin`:
 
-### Common Output Fields
+```typescript
+{
+  "session_id": string,      // Unique ID for the current session
+  "transcript_path": string, // Absolute path to session transcript JSON
+  "cwd": string,             // Current working directory
+  "hook_event_name": string, // The firing event (e.g. "BeforeTool")
+  "timestamp": string        // ISO 8601 execution time
+}
+```
 
-| Field                | Type      | Description                                                              |
-| :------------------- | :-------- | :----------------------------------------------------------------------- |
-| `decision`           | `string`  | One of: `allow`, `deny`, `block`, `ask`, `approve`.                      |
-| `reason`             | `string`  | Explanation shown to the **agent** when a decision is `deny` or `block`. |
-| `systemMessage`      | `string`  | Message displayed to the **user** in the CLI terminal.                   |
-| `continue`           | `boolean` | If `false`, immediately terminates the agent loop for this turn.         |
-| `stopReason`         | `string`  | Message shown to the user when `continue` is `false`.                    |
-| `suppressOutput`     | `boolean` | If `true`, the hook execution is hidden from the CLI transcript.         |
-| `hookSpecificOutput` | `object`  | Container for event-specific data (see below).                           |
+---
 
-### `hookSpecificOutput` Reference
+## Common output fields
 
-| Field               | Supported Events                           | Description                                                                       |
-| :------------------ | :----------------------------------------- | :-------------------------------------------------------------------------------- |
-| `additionalContext` | `SessionStart`, `BeforeAgent`, `AfterTool` | Appends text directly to the agent's context.                                     |
-| `llm_request`       | `BeforeModel`                              | A `Partial<LLMRequest>` to override parameters of the outgoing call.              |
-| `llm_response`      | `BeforeModel`                              | A **full** `LLMResponse` to bypass the model and provide a synthetic result.      |
-| `llm_response`      | `AfterModel`                               | A `Partial<LLMResponse>` to modify the model's response before the agent sees it. |
-| `toolConfig`        | `BeforeToolSelection`                      | Object containing `mode` (`AUTO`/`ANY`/`NONE`) and `allowedFunctionNames`.        |
+Most hooks support these fields in their `stdout` JSON:
+
+| Field            | Type      | Description                                                                    |
+| :--------------- | :-------- | :----------------------------------------------------------------------------- |
+| `systemMessage`  | `string`  | Displayed immediately to the user in the terminal.                             |
+| `suppressOutput` | `boolean` | If `true`, hides internal hook metadata from logs/telemetry.                   |
+| `continue`       | `boolean` | If `false`, stops the entire agent loop immediately.                           |
+| `stopReason`     | `string`  | Displayed to the user when `continue` is `false`.                              |
+| `decision`       | `string`  | `"allow"` or `"deny"` (alias `"block"`). Specific impact depends on the event. |
+| `reason`         | `string`  | The feedback/error message provided when a `decision` is `"deny"`.             |
+
+---
+
+## Tool hooks
+
+### Matchers and tool names
+
+For `BeforeTool` and `AfterTool` events, the `matcher` field in your settings is
+compared against the name of the tool being executed.
+
+- **Built-in Tools**: You can match any built-in tool (e.g., `read_file`,
+  `run_shell_command`). See the [Tools Reference](/docs/tools) for a full list
+  of available tool names.
+- **MCP Tools**: Tools from MCP servers follow the naming pattern
+  `mcp__<server_name>__<tool_name>`.
+- **Regex Support**: Matchers support regular expressions (e.g.,
+  `matcher: "read_.*"` matches all file reading tools).
+
+### `BeforeTool`
+
+Fires before a tool is invoked. Used for argument validation, security checks,
+and parameter rewriting.
+
+- **Input Fields**:
+  - `tool_name`: (`string`) The name of the tool being called.
+  - `tool_input`: (`object`) The raw arguments generated by the model.
+  - `mcp_context`: (`object`) Optional metadata for MCP-based tools.
+- **Relevant Output Fields**:
+  - `decision`: Set to `"deny"` (or `"block"`) to prevent the tool from
+    executing.
+  - `reason`: Required if denied. This text is sent **to the agent** as a tool
+    error, allowing it to respond or retry.
+  - `hookSpecificOutput.tool_input`: An object that **merges with and
+    overrides** the model's arguments before execution.
+  - `continue`: Set to `false` to **kill the entire agent loop** immediately.
+- **Exit Code 2 (Block Tool)**: Prevents execution. Uses `stderr` as the
+  `reason` sent to the agent. **The turn continues.**
+
+### `AfterTool`
+
+Fires after a tool executes. Used for result auditing, context injection, or
+hiding sensitive output from the agent.
+
+- **Input Fields**:
+  - `tool_name`: (`string`)
+  - `tool_input`: (`object`) The original arguments.
+  - `tool_response`: (`object`) The result containing `llmContent`,
+    `returnDisplay`, and optional `error`.
+  - `mcp_context`: (`object`)
+- **Relevant Output Fields**:
+  - `decision`: Set to `"deny"` to hide the real tool output from the agent.
+  - `reason`: Required if denied. This text **replaces** the tool result sent
+    back to the model.
+  - `hookSpecificOutput.additionalContext`: Text that is **appended** to the
+    tool result for the agent.
+  - `continue`: Set to `false` to **kill the entire agent loop** immediately.
+- **Exit Code 2 (Block Result)**: Hides the tool result. Uses `stderr` as the
+  replacement content sent to the agent. **The turn continues.**
+
+---
+
+## Agent hooks
+
+### `BeforeAgent`
+
+Fires after a user submits a prompt, but before the agent begins planning. Used
+for prompt validation or injecting dynamic context.
+
+- **Input Fields**:
+  - `prompt`: (`string`) The original text submitted by the user.
+- **Relevant Output Fields**:
+  - `hookSpecificOutput.additionalContext`: Text that is **appended** to the
+    prompt for this turn only.
+  - `decision`: Set to `"deny"` to block the turn and **discard the user's
+    message** (it will not appear in history).
+  - `continue`: Set to `false` to block the turn but **save the message to
+    history**.
+  - `reason`: Required if denied or stopped.
+- **Exit Code 2 (Block Turn)**: Aborts the turn and erases the prompt from
+  context. Same as `decision: "deny"`.
+
+### `AfterAgent`
+
+Fires once per turn after the model generates its final response. Primary use
+case is response validation and automatic retries.
+
+- **Input Fields**:
+  - `prompt`: (`string`) The user's original request.
+  - `prompt_response`: (`string`) The final text generated by the agent.
+  - `stop_hook_active`: (`boolean`) Indicates if this hook is already running as
+    part of a retry sequence.
+- **Relevant Output Fields**:
+  - `decision`: Set to `"deny"` to **reject the response** and force a retry.
+  - `reason`: Required if denied. This text is sent **to the agent as a new
+    prompt** to request a correction.
+  - `continue`: Set to `false` to **stop the session** without retrying.
+  - `clearContext`: If `true`, clears conversation history (LLM memory) while
+    preserving UI display.
+- **Exit Code 2 (Retry)**: Rejects the response and triggers an automatic retry
+  turn using `stderr` as the feedback prompt.
+
+---
+
+## Model hooks
+
+### `BeforeModel`
+
+Fires before sending a request to the LLM. Operates on a stable, SDK-agnostic
+request format.
+
+- **Input Fields**:
+  - `llm_request`: (`object`) Contains `model`, `messages`, and `config`
+    (generation params).
+- **Relevant Output Fields**:
+  - `hookSpecificOutput.llm_request`: An object that **overrides** parts of the
+    outgoing request (e.g., changing models or temperature).
+  - `hookSpecificOutput.llm_response`: A **Synthetic Response** object. If
+    provided, the CLI skips the LLM call entirely and uses this as the response.
+  - `decision`: Set to `"deny"` to block the request and abort the turn.
+- **Exit Code 2 (Block Turn)**: Aborts the turn and skips the LLM call. Uses
+  `stderr` as the error message.
+
+### `BeforeToolSelection`
+
+Fires before the LLM decides which tools to call. Used to filter the available
+toolset or force specific tool modes.
+
+- **Input Fields**:
+  - `llm_request`: (`object`) Same format as `BeforeModel`.
+- **Relevant Output Fields**:
+  - `hookSpecificOutput.toolConfig.mode`: (`"AUTO" | "ANY" | "NONE"`)
+    - `"NONE"`: Disables all tools (Wins over other hooks).
+    - `"ANY"`: Forces at least one tool call.
+  - `hookSpecificOutput.toolConfig.allowedFunctionNames`: (`string[]`) Whitelist
+    of tool names.
+- **Union Strategy**: Multiple hooks' whitelists are **combined**.
+- **Limitations**: Does **not** support `decision`, `continue`, or
+  `systemMessage`.
+
+### `AfterModel`
+
+Fires immediately after an LLM response chunk is received. Used for real-time
+redaction or PII filtering.
+
+- **Input Fields**:
+  - `llm_request`: (`object`) The original request.
+  - `llm_response`: (`object`) The model's response (or a single chunk during
+    streaming).
+- **Relevant Output Fields**:
+  - `hookSpecificOutput.llm_response`: An object that **replaces** the model's
+    response chunk.
+  - `decision`: Set to `"deny"` to discard the response chunk and block the
+    turn.
+  - `continue`: Set to `false` to **kill the entire agent loop** immediately.
+- **Note on Streaming**: Fired for **every chunk** generated by the model.
+  Modifying the response only affects the current chunk.
+- **Exit Code 2 (Block Response)**: Aborts the turn and discards the model's
+  output. Uses `stderr` as the error message.
+
+---
+
+## Lifecycle & system hooks
+
+### `SessionStart`
+
+Fires on application startup, resuming a session, or after a `/clear` command.
+Used for loading initial context.
+
+- **Input fields**:
+  - `source`: (`"startup" | "resume" | "clear"`)
+- **Relevant output fields**:
+  - `hookSpecificOutput.additionalContext`: (`string`)
+    - **Interactive**: Injected as the first turn in history.
+    - **Non-interactive**: Prepended to the user's prompt.
+  - `systemMessage`: Shown at the start of the session.
+- **Advisory only**: `continue` and `decision` fields are **ignored**. Startup
+  is never blocked.
+
+### `SessionEnd`
+
+Fires when the CLI exits or a session is cleared. Used for cleanup or final
+telemetry.
+
+- **Input Fields**:
+  - `reason`: (`"exit" | "clear" | "logout" | "prompt_input_exit" | "other"`)
+- **Relevant Output Fields**:
+  - `systemMessage`: Displayed to the user during shutdown.
+- **Best Effort**: The CLI **will not wait** for this hook to complete and
+  ignores all flow-control fields (`continue`, `decision`).
+
+### `Notification`
+
+Fires when the CLI emits a system alert (e.g., Tool Permissions). Used for
+external logging or cross-platform alerts.
+
+- **Input Fields**:
+  - `notification_type`: (`"ToolPermission"`)
+  - `message`: Summary of the alert.
+  - `details`: JSON object with alert-specific metadata (e.g., tool name, file
+    path).
+- **Relevant Output Fields**:
+  - `systemMessage`: Displayed alongside the system alert.
+- **Observability Only**: This hook **cannot** block alerts or grant permissions
+  automatically. Flow-control fields are ignored.
+
+### `PreCompress`
+
+Fires before the CLI summarizes history to save tokens. Used for logging or
+state saving.
+
+- **Input Fields**:
+  - `trigger`: (`"auto" | "manual"`)
+- **Relevant Output Fields**:
+  - `systemMessage`: Displayed to the user before compression.
+- **Advisory Only**: Fired asynchronously. It **cannot** block or modify the
+  compression process. Flow-control fields are ignored.
 
 ---
 
 ## Stable Model API
 
-Gemini CLI uses a decoupled format for model interactions to ensure hooks remain
-stable even if the underlying Gemini SDK changes.
+Gemini CLI uses these structures to ensure hooks don't break across SDK updates.
 
-### `LLMRequest` Object
-
-Used in `BeforeModel` and `BeforeToolSelection`.
-
-> 💡 **Note**: In v1, model hooks are primarily text-focused. Non-text parts
-> (like images or function calls) provided in the `content` array will be
-> simplified to their string representation by the translator.
+**LLMRequest**:
 
 ```typescript
 {
   "model": string,
   "messages": Array<{
     "role": "user" | "model" | "system",
-    "content": string | Array<{ "type": string, [key: string]: any }>
+    "content": string // Non-text parts are filtered out for hooks
   }>,
-  "config"?: {
-    "temperature"?: number,
-    "maxOutputTokens"?: number,
-    "topP"?: number,
-    "topK"?: number
-  },
-  "toolConfig"?: {
-    "mode"?: "AUTO" | "ANY" | "NONE",
-    "allowedFunctionNames"?: string[]
-  }
+  "config": { "temperature": number, ... },
+  "toolConfig": { "mode": string, "allowedFunctionNames": string[] }
 }
+
 ```
 
-### `LLMResponse` Object
-
-Used in `AfterModel` and as a synthetic response in `BeforeModel`.
+**LLMResponse**:
 
 ```typescript
 {
-  "text"?: string,
   "candidates": Array<{
-    "content": {
-      "role": "model",
-      "parts": string[]
-    },
-    "finishReason"?: "STOP" | "MAX_TOKENS" | "SAFETY" | "RECITATION" | "OTHER",
-    "index"?: number,
-    "safetyRatings"?: Array<{
-      "category": string,
-      "probability": string,
-      "blocked"?: boolean
-    }>
+    "content": { "role": "model", "parts": string[] },
+    "finishReason": string
   }>,
-  "usageMetadata"?: {
-    "promptTokenCount"?: number,
-    "candidatesTokenCount"?: number,
-    "totalTokenCount"?: number
-  }
+  "usageMetadata": { "totalTokenCount": number }
 }
 ```

docs/index.md

@@ -56,6 +56,8 @@ This documentation is organized into the following sections:
   commands with `/model`.
 - **[Sandbox](./cli/sandbox.md):** Isolate tool execution in a secure,
   containerized environment.
+- **[Agent Skills](./cli/skills.md):** (Experimental) Extend the CLI with
+  specialized expertise and procedural workflows.
 - **[Settings](./cli/settings.md):** Configure various aspects of the CLI's
   behavior and appearance with `/settings`.
 - **[Telemetry](./cli/telemetry.md):** Overview of telemetry in the CLI.
@@ -100,10 +102,10 @@ This documentation is organized into the following sections:
 
 - **[Introduction: Extensions](./extensions/index.md):** How to extend the CLI
   with new functionality.
-- **[Get Started with extensions](./extensions/getting-started-extensions.md):**
-  Learn how to build your own extension.
-- **[Extension releasing](./extensions/extension-releasing.md):** How to release
-  Gemini CLI extensions.
+- **[Writing extensions](./extensions/writing-extensions.md):** Learn how to
+  build your own extension.
+- **[Extension releasing](./extensions/releasing.md):** How to release Gemini
+  CLI extensions.
 
 ### Hooks
 

docs/local-development.md

@@ -10,7 +10,7 @@ debug your code by instrumenting interesting events like model calls, tool
 scheduler, tool calls, etc.
 
 Dev traces are verbose and are specifically meant for understanding agent
-behaviour and debugging issues. They are disabled by default.
+behavior and debugging issues. They are disabled by default.
 
 To enable dev traces, set the `GEMINI_DEV_TRACING=true` environment variable
 when running Gemini CLI.

docs/releases.md

@@ -12,7 +12,7 @@ Dressing Room, which is Google's system for managing NPM packages in the
 `@google/**` namespace. The packages are all named `@google/**`.
 
 More information can be found about these systems in the
-[maintainer repo guide](https://github.com/google-gemini/maintainers-gemini-cli/blob/main/npm.md)
+[NPM Package Overview](npm.md)
 
 ### Package scopes
 

docs/sidebar.json

@@ -1,187 +1,52 @@
 [
-  {
-    "label": "Overview",
-    "items": [
-      {
-        "label": "Introduction",
-        "slug": "docs"
-      },
-      {
-        "label": "Architecture overview",
-        "slug": "docs/architecture"
-      },
-      {
-        "label": "Contribution guide",
-        "slug": "docs/contributing"
-      }
-    ]
-  },
   {
     "label": "Get started",
     "items": [
-      {
-        "label": "Gemini CLI quickstart",
-        "slug": "docs/get-started"
-      },
-      {
-        "label": "Gemini 3 on Gemini CLI",
-        "slug": "docs/get-started/gemini-3"
-      },
-      {
-        "label": "Authentication",
-        "slug": "docs/get-started/authentication"
-      },
-      {
-        "label": "Configuration",
-        "slug": "docs/get-started/configuration"
-      },
-      {
-        "label": "Installation",
-        "slug": "docs/get-started/installation"
-      },
-      {
-        "label": "Examples",
-        "slug": "docs/get-started/examples"
-      }
+      { "label": "Overview", "slug": "docs" },
+      { "label": "Quickstart", "slug": "docs/get-started" },
+      { "label": "Installation", "slug": "docs/get-started/installation" },
+      { "label": "Authentication", "slug": "docs/get-started/authentication" },
+      { "label": "Examples", "slug": "docs/get-started/examples" },
+      { "label": "Gemini 3 (preview)", "slug": "docs/get-started/gemini-3" }
     ]
   },
   {
-    "label": "CLI",
+    "label": "Use Gemini CLI",
     "items": [
-      {
-        "label": "Introduction",
-        "slug": "docs/cli"
-      },
-      {
-        "label": "Commands",
-        "slug": "docs/cli/commands"
-      },
-      {
-        "label": "Checkpointing",
-        "slug": "docs/cli/checkpointing"
-      },
-      {
-        "label": "Custom commands",
-        "slug": "docs/cli/custom-commands"
-      },
-      {
-        "label": "Enterprise",
-        "slug": "docs/cli/enterprise"
-      },
-      {
-        "label": "Headless mode",
-        "slug": "docs/cli/headless"
-      },
-      {
-        "label": "Keyboard shortcuts",
-        "slug": "docs/cli/keyboard-shortcuts"
-      },
-      {
-        "label": "Model selection",
-        "slug": "docs/cli/model"
-      },
-      {
-        "label": "Sandbox",
-        "slug": "docs/cli/sandbox"
-      },
-      {
-        "label": "Session Management",
-        "slug": "docs/cli/session-management"
-      },
-      {
-        "label": "Agent Skills",
-        "slug": "docs/cli/skills"
-      },
-      {
-        "label": "Settings",
-        "slug": "docs/cli/settings"
-      },
-      {
-        "label": "Telemetry",
-        "slug": "docs/cli/telemetry"
-      },
-      {
-        "label": "Themes",
-        "slug": "docs/cli/themes"
-      },
-      {
-        "label": "Token caching",
-        "slug": "docs/cli/token-caching"
-      },
-      {
-        "label": "Trusted Folders",
-        "slug": "docs/cli/trusted-folders"
-      },
-      {
-        "label": "Tutorials",
-        "slug": "docs/cli/tutorials"
-      },
-      {
-        "label": "Uninstall",
-        "slug": "docs/cli/uninstall"
-      },
-      {
-        "label": "System prompt override",
-        "slug": "docs/cli/system-prompt"
-      }
+      { "label": "Using the CLI", "slug": "docs/cli" },
+      { "label": "File management", "slug": "docs/tools/file-system" },
+      { "label": "Memory management", "slug": "docs/tools/memory" },
+      { "label": "Project context (GEMINI.md)", "slug": "docs/cli/gemini-md" },
+      { "label": "Shell commands", "slug": "docs/tools/shell" },
+      { "label": "Session management", "slug": "docs/cli/session-management" },
+      { "label": "Todos", "slug": "docs/tools/todos" },
+      { "label": "Web search and fetch", "slug": "docs/tools/web-search" }
     ]
   },
   {
-    "label": "Core",
+    "label": "Configuration",
     "items": [
       {
-        "label": "Introduction",
-        "slug": "docs/core"
+        "label": "Ignore files (.geminiignore)",
+        "slug": "docs/cli/gemini-ignore"
       },
-      {
-        "label": "Tools API",
-        "slug": "docs/core/tools-api"
-      },
-      {
-        "label": "Memory Import Processor",
-        "slug": "docs/core/memport"
-      },
-      {
-        "label": "Policy Engine",
-        "slug": "docs/core/policy-engine"
-      }
+      { "label": "Model selection", "slug": "docs/cli/model" },
+      { "label": "Settings", "slug": "docs/cli/settings" },
+      { "label": "Themes", "slug": "docs/cli/themes" },
+      { "label": "Token caching", "slug": "docs/cli/token-caching" },
+      { "label": "Trusted folders", "slug": "docs/cli/trusted-folders" }
     ]
   },
   {
-    "label": "Tools",
+    "label": "Advanced features",
     "items": [
-      {
-        "label": "Introduction",
-        "slug": "docs/tools"
-      },
-      {
-        "label": "File system",
-        "slug": "docs/tools/file-system"
-      },
-      {
-        "label": "Shell",
-        "slug": "docs/tools/shell"
-      },
-      {
-        "label": "Web fetch",
-        "slug": "docs/tools/web-fetch"
-      },
-      {
-        "label": "Web search",
-        "slug": "docs/tools/web-search"
-      },
-      {
-        "label": "Memory",
-        "slug": "docs/tools/memory"
-      },
-      {
-        "label": "Todos",
-        "slug": "docs/tools/todos"
-      },
-      {
-        "label": "MCP servers",
-        "slug": "docs/tools/mcp-server"
-      }
+      { "label": "Checkpointing", "slug": "docs/cli/checkpointing" },
+      { "label": "Custom commands", "slug": "docs/cli/custom-commands" },
+      { "label": "Enterprise features", "slug": "docs/cli/enterprise" },
+      { "label": "Headless mode & scripting", "slug": "docs/cli/headless" },
+      { "label": "Sandboxing", "slug": "docs/cli/sandbox" },
+      { "label": "System prompt override", "slug": "docs/cli/system-prompt" },
+      { "label": "Telemetry", "slug": "docs/cli/telemetry" }
     ]
   },
   {
@@ -192,110 +57,76 @@
         "slug": "docs/extensions"
       },
       {
-        "label": "Get started with extensions",
-        "slug": "docs/extensions/getting-started-extensions"
+        "label": "Writing extensions",
+        "slug": "docs/extensions/writing-extensions"
       },
       {
-        "label": "Extension releasing",
-        "slug": "docs/extensions/extension-releasing"
-      }
-    ]
-  },
-  {
-    "label": "Hooks",
-    "items": [
-      {
-        "label": "Introduction",
-        "slug": "docs/hooks"
-      },
-      {
-        "label": "Writing hooks",
-        "slug": "docs/hooks/writing-hooks"
-      },
-      {
-        "label": "Hooks reference",
-        "slug": "docs/hooks/reference"
+        "label": "Extensions reference",
+        "slug": "docs/extensions/reference"
       },
       {
         "label": "Best practices",
-        "slug": "docs/hooks/best-practices"
+        "slug": "docs/extensions/best-practices"
+      },
+      {
+        "label": "Extensions releasing",
+        "slug": "docs/extensions/releasing"
       }
     ]
   },
   {
-    "label": "IDE integration",
+    "label": "Ecosystem and extensibility",
+    "items": [
+      { "label": "Agent skills (experimental)", "slug": "docs/cli/skills" },
+      { "label": "Hooks (experimental)", "slug": "docs/hooks" },
+      { "label": "IDE integration", "slug": "docs/ide-integration" },
+      { "label": "MCP servers", "slug": "docs/tools/mcp-server" }
+    ]
+  },
+  {
+    "label": "Tutorials",
     "items": [
       {
-        "label": "Introduction",
-        "slug": "docs/ide-integration"
+        "label": "Get started with extensions",
+        "slug": "docs/extensions/writing-extensions"
       },
-      {
-        "label": "IDE companion spec",
-        "slug": "docs/ide-integration/ide-companion-spec"
-      }
+      { "label": "How to write hooks", "slug": "docs/hooks/writing-hooks" }
+    ]
+  },
+  {
+    "label": "Reference",
+    "items": [
+      { "label": "Architecture", "slug": "docs/architecture" },
+      { "label": "Command reference", "slug": "docs/cli/commands" },
+      { "label": "Configuration", "slug": "docs/get-started/configuration" },
+      { "label": "Keyboard shortcuts", "slug": "docs/cli/keyboard-shortcuts" },
+      { "label": "Memory import processor", "slug": "docs/core/memport" },
+      { "label": "Policy engine", "slug": "docs/core/policy-engine" },
+      { "label": "Tools API", "slug": "docs/core/tools-api" }
+    ]
+  },
+  {
+    "label": "Resources",
+    "items": [
+      { "label": "FAQ", "slug": "docs/faq" },
+      { "label": "Quota and pricing", "slug": "docs/quota-and-pricing" },
+      { "label": "Release notes", "slug": "docs/changelogs/" },
+      { "label": "Terms and privacy", "slug": "docs/tos-privacy" },
+      { "label": "Troubleshooting", "slug": "docs/troubleshooting" },
+      { "label": "Uninstall", "slug": "docs/cli/uninstall" }
     ]
   },
   {
     "label": "Development",
     "items": [
-      {
-        "label": "NPM",
-        "slug": "docs/npm"
-      },
-      {
-        "label": "Releases",
-        "slug": "docs/releases"
-      },
-      {
-        "label": "Integration tests",
-        "slug": "docs/integration-tests"
-      },
+      { "label": "Contribution guide", "slug": "docs/contributing" },
+      { "label": "Integration testing", "slug": "docs/integration-tests" },
       {
         "label": "Issue and PR automation",
         "slug": "docs/issue-and-pr-automation"
-      }
-    ]
-  },
-  {
-    "label": "Releases",
-    "items": [
-      {
-        "label": "Release notes",
-        "slug": "docs/changelogs/"
       },
-      {
-        "label": "Latest release",
-        "slug": "docs/changelogs/latest"
-      },
-      {
-        "label": "Preview release",
-        "slug": "docs/changelogs/preview"
-      },
-      {
-        "label": "Changelog",
-        "slug": "docs/changelogs/releases"
-      }
-    ]
-  },
-  {
-    "label": "Support",
-    "items": [
-      {
-        "label": "FAQ",
-        "slug": "docs/faq"
-      },
-      {
-        "label": "Troubleshooting",
-        "slug": "docs/troubleshooting"
-      },
-      {
-        "label": "Quota and pricing",
-        "slug": "docs/quota-and-pricing"
-      },
-      {
-        "label": "Terms of service",
-        "slug": "docs/tos-privacy"
-      }
+      { "label": "Local development", "slug": "docs/local-development" },
+      { "label": "NPM package structure", "slug": "docs/npm" }
     ]
   }
 ]

docs/tools/index.md

@@ -91,5 +91,8 @@ Additionally, these tools incorporate:
 
 - **[MCP servers](./mcp-server.md)**: MCP servers act as a bridge between the
   Gemini model and your local environment or other services like APIs.
+- **[Agent Skills](../cli/skills.md)**: (Experimental) On-demand expertise
+  packages that are activated via the `activate_skill` tool to provide
+  specialized guidance and resources.
 - **[Sandboxing](../cli/sandbox.md)**: Sandboxing isolates the model and its
   changes from your environment to reduce potential risk.

docs/tools/mcp-server.md

@@ -722,7 +722,8 @@ The MCP integration tracks several states:
 
 ### Debugging tips
 
-1. **Enable debug mode:** Run the CLI with `--debug` for verbose output
+1. **Enable debug mode:** Run the CLI with `--debug` for verbose output (use F12
+   to open debug console in interactive mode)
 2. **Check stderr:** MCP server stderr is captured and logged (INFO messages
    filtered)
 3. **Test isolation:** Test your MCP server independently before integrating
@@ -732,7 +733,7 @@ The MCP integration tracks several states:
 
 ## Important notes
 
-### Security sonsiderations
+### Security considerations
 
 - **Trust settings:** The `trust` option bypasses all confirmation dialogs. Use
   cautiously and only for servers you completely control
@@ -1037,6 +1038,29 @@ gemini mcp remove my-server
 This will find and delete the "my-server" entry from the `mcpServers` object in
 the appropriate `settings.json` file based on the scope (`-s, --scope`).
 
+### Enabling/disabling a server (`gemini mcp enable`, `gemini mcp disable`)
+
+Temporarily disable an MCP server without removing its configuration, or
+re-enable a previously disabled server.
+
+**Commands:**
+
+```bash
+gemini mcp enable <name> [--session]
+gemini mcp disable <name> [--session]
+```
+
+**Options (flags):**
+
+- `--session`: Apply change only for this session (not persisted to file).
+
+Disabled servers appear in `/mcp` status as "Disabled" but won't connect or
+provide tools. Enablement state is stored in
+`~/.gemini/mcp-server-enablement.json`.
+
+The same commands are available as slash commands during an active session:
+`/mcp enable <name>` and `/mcp disable <name>`.
+
 ## Instructions
 
 Gemini CLI supports

docs/tools/shell.md

@@ -135,7 +135,7 @@ user input, such as text editors (`vim`, `nano`), terminal-based UIs (`htop`),
 and interactive version control operations (`git rebase -i`).
 
 When an interactive command is running, you can send input to it from the Gemini
-CLI. To focus on the interactive shell, press `ctrl+f`. The terminal output,
+CLI. To focus on the interactive shell, press `Tab`. The terminal output,
 including complex TUIs, will be rendered correctly.
 
 ## Important notes

docs/troubleshooting.md

@@ -28,6 +28,13 @@ topics on:
     - **Organizational Users:** Contact your Google Cloud administrator to be
       added to your organization's Gemini Code Assist subscription.
 
+- **Error:
+  `Failed to login. Message: Your current account is not eligible... because it is not currently available in your location.`**
+  - **Cause:** Gemini CLI does not currently support your location. For a full
+    list of supported locations, see the following pages:
+    - Gemini Code Assist for individuals:
+      [Available locations](https://developers.google.com/gemini-code-assist/resources/available-locations#americas)
+
 - **Error: `Failed to login. Message: Request contains an invalid argument`**
   - **Cause:** Users with Google Workspace accounts or Google Cloud accounts
     associated with their Gmail accounts may not be able to activate the free
@@ -130,13 +137,15 @@ This is especially useful for scripting and automation.
 ## Debugging tips
 
 - **CLI debugging:**
-  - Use the `--debug` flag for more detailed output.
+  - Use the `--debug` flag for more detailed output. In interactive mode, press
+    F12 to view the debug console.
   - Check the CLI logs, often found in a user-specific configuration or cache
     directory.
 
 - **Core debugging:**
   - Check the server console output for error messages or stack traces.
-  - Increase log verbosity if configurable.
+  - Increase log verbosity if configurable. For example, set the `DEBUG_MODE`
+    environment variable to `true` or `1`.
   - Use Node.js debugging tools (e.g., `node --inspect`) if you need to step
     through server-side code.
 

esbuild.config.js

@@ -14,8 +14,8 @@ let esbuild;
 try {
   esbuild = (await import('esbuild')).default;
 } catch (_error) {
-  console.warn('esbuild not available, skipping bundle step');
-  process.exit(0);
+  console.error('esbuild not available - cannot build bundle');
+  process.exit(1);
 }
 
 const __filename = fileURLToPath(import.meta.url);

eslint.config.js

@@ -35,6 +35,9 @@ export default tseslint.config(
       'package/bundle/**',
       '.integration-tests/**',
       'dist/**',
+      'evals/**',
+      'packages/test-utils/**',
+      'packages/core/src/skills/builtin/skill-creator/scripts/*.cjs',
     ],
   },
   eslint.configs.recommended,
@@ -301,6 +304,16 @@ export default tseslint.config(
       '@typescript-eslint/no-require-imports': 'off',
     },
   },
+  // Examples should have access to standard globals like fetch
+  {
+    files: ['packages/cli/src/commands/extensions/examples/**/*.js'],
+    languageOptions: {
+      globals: {
+        ...globals.node,
+        fetch: 'readonly',
+      },
+    },
+  },
   // extra settings for scripts that we run directly with node
   {
     files: ['packages/vscode-ide-companion/scripts/**/*.js'],

evals/README.md

@@ -0,0 +1,191 @@
+# Behavioral Evals
+
+Behavioral evaluations (evals) are tests designed to validate the agent's
+behavior in response to specific prompts. They serve as a critical feedback loop
+for changes to system prompts, tool definitions, and other model-steering
+mechanisms.
+
+## Why Behavioral Evals?
+
+Unlike traditional **integration tests** which verify that the system functions
+correctly (e.g., "does the file writer actually write to disk?"), behavioral
+evals verify that the model _chooses_ to take the correct action (e.g., "does
+the model decide to write to disk when asked to save code?").
+
+They are also distinct from broad **industry benchmarks** (like SWE-bench).
+While benchmarks measure general capabilities across complex challenges, our
+behavioral evals focus on specific, granular behaviors relevant to the Gemini
+CLI's features.
+
+### Key Characteristics
+
+- **Feedback Loop**: They help us understand how changes to prompts or tools
+  affect the model's decision-making.
+  - _Did a change to the system prompt make the model less likely to use tool
+    X?_
+  - _Did a new tool definition confuse the model?_
+- **Regression Testing**: They prevent regressions in model steering.
+- **Non-Determinism**: Unlike unit tests, LLM behavior can be non-deterministic.
+  We distinguish between behaviors that should be robust (`ALWAYS_PASSES`) and
+  those that are generally reliable but might occasionally vary
+  (`USUALLY_PASSES`).
+
+## Creating an Evaluation
+
+Evaluations are located in the `evals` directory. Each evaluation is a Vitest
+test file that uses the `evalTest` function from `evals/test-helper.ts`.
+
+### `evalTest`
+
+The `evalTest` function is a helper that runs a single evaluation case. It takes
+two arguments:
+
+1. `policy`: The consistency expectation for this test (`'ALWAYS_PASSES'` or
+   `'USUALLY_PASSES'`).
+2. `evalCase`: An object defining the test case.
+
+#### Policies
+
+Policies control how strictly a test is validated. Tests should generally use
+the ALWAYS_PASSES policy to offer the strictest guarantees.
+
+USUALLY_PASSES exists to enable assertion of less consistent or aspirational
+behaviors.
+
+- `ALWAYS_PASSES`: Tests expected to pass 100% of the time. These are typically
+  trivial and test basic functionality. These run in every CI.
+- `USUALLY_PASSES`: Tests expected to pass most of the time but may have some
+  flakiness due to non-deterministic behaviors. These are run nightly and used
+  to track the health of the product from build to build.
+
+#### `EvalCase` Properties
+
+- `name`: The name of the evaluation case.
+- `prompt`: The prompt to send to the model.
+- `params`: An optional object with parameters to pass to the test rig (e.g.,
+  settings).
+- `assert`: An async function that takes the test rig and the result of the run
+  and asserts that the result is correct.
+- `log`: An optional boolean that, if set to `true`, will log the tool calls to
+  a file in the `evals/logs` directory.
+
+### Example
+
+```typescript
+import { describe, expect } from 'vitest';
+import { evalTest } from './test-helper.js';
+
+describe('my_feature', () => {
+  evalTest('ALWAYS_PASSES', {
+    name: 'should do something',
+    prompt: 'do it',
+    assert: async (rig, result) => {
+      // assertions
+    },
+  });
+});
+```
+
+## Running Evaluations
+
+First, build the bundled Gemini CLI. You must do this after every code change.
+
+```bash
+npm run build
+npm run bundle
+```
+
+### Always Passing Evals
+
+To run the evaluations that are expected to always pass (CI safe):
+
+```bash
+npm run test:always_passing_evals
+```
+
+### All Evals
+
+To run all evaluations, including those that may be flaky ("usually passes"):
+
+```bash
+npm run test:all_evals
+```
+
+This command sets the `RUN_EVALS` environment variable to `1`, which enables the
+`USUALLY_PASSES` tests.
+
+## Reporting
+
+Results for evaluations are available on GitHub Actions:
+
+- **CI Evals**: Included in the
+  [E2E (Chained)](https://github.com/google-gemini/gemini-cli/actions/workflows/chained_e2e.yml)
+  workflow. These must pass 100% for every PR.
+- **Nightly Evals**: Run daily via the
+  [Evals: Nightly](https://github.com/google-gemini/gemini-cli/actions/workflows/evals-nightly.yml)
+  workflow. These track the long-term health and stability of model steering.
+
+### Nightly Report Format
+
+The nightly workflow executes the full evaluation suite multiple times
+(currently 3 attempts) to account for non-determinism. These results are
+aggregated into a **Nightly Summary** attached to the workflow run.
+
+#### How to interpret the report:
+
+- **Pass Rate (%)**: Each cell represents the percentage of successful runs for
+  a specific test in that workflow instance.
+- **History**: The table shows the pass rates for the last 10 nightly runs,
+  allowing you to identify if a model's behavior is trending towards
+  instability.
+- **Total Pass Rate**: An aggregate metric of all evaluations run in that batch.
+
+A significant drop in the pass rate for a `USUALLY_PASSES` test—even if it
+doesn't drop to 0%—often indicates that a recent change to a system prompt or
+tool definition has made the model's behavior less reliable.
+
+## Fixing Evaluations
+
+If an evaluation is failing or has a regressed pass rate, you can use the
+`/fix-behavioral-eval` command within Gemini CLI to help investigate and fix the
+issue.
+
+### `/fix-behavioral-eval`
+
+This command is designed to automate the investigation and fixing process for
+failing evaluations. It will:
+
+1.  **Investigate**: Fetch the latest results from the nightly workflow using
+    the `gh` CLI, identify the failing test, and review test trajectory logs in
+    `evals/logs`.
+2.  **Fix**: Suggest and apply targeted fixes to the prompt or tool definitions.
+    It prioritizes minimal changes to `prompt.ts`, tool instructions, and
+    modules that contribute to the prompt. It generally tries to avoid changing
+    the test itself.
+3.  **Verify**: Re-run the test 3 times across multiple models (e.g., Gemini
+    3.0, Gemini 3 Flash, Gemini 2.5 Pro) to ensure stability and calculate a
+    success rate.
+4.  **Report**: Provide a summary of the success rate for each model and details
+    on the applied fixes.
+
+To use it, run:
+
+```bash
+gemini /fix-behavioral-eval
+```
+
+You can also provide a link to a specific GitHub Action run or the name of a
+specific test to focus the investigation:
+
+```bash
+gemini /fix-behavioral-eval https://github.com/google-gemini/gemini-cli/actions/runs/123456789
+```
+
+When investigating failures manually, you can also enable verbose agent logs by
+setting the `GEMINI_DEBUG_LOG_FILE` environment variable.
+
+It's highly recommended to manually review and/or ask the agent to iterate on
+any prompt changes, even if they pass all evals. The prompt should prefer
+positive traits ('do X') and resort to negative traits ('do not do X') only when
+unable to accomplish the goal with positive traits. Gemini is quite good at
+instrospecting on its prompt when asked the right questions.

evals/generalist_agent.eval.ts

@@ -0,0 +1,41 @@
+/**
+ * @license
+ * Copyright 2025 Google LLC
+ * SPDX-License-Identifier: Apache-2.0
+ */
+
+import { describe, expect } from 'vitest';
+import { evalTest } from './test-helper.js';
+import path from 'node:path';
+import fs from 'node:fs/promises';
+
+describe('generalist_agent', () => {
+  evalTest('USUALLY_PASSES', {
+    name: 'should be able to use generalist agent by explicitly asking the main agent to invoke it',
+    params: {
+      settings: {
+        agents: {
+          overrides: {
+            generalist: { enabled: true },
+          },
+        },
+      },
+    },
+    prompt:
+      'Please use the generalist agent to create a file called "generalist_test_file.txt" containing exactly the following text: success',
+    assert: async (rig) => {
+      // 1) Verify the generalist agent was invoked via delegate_to_agent
+      const foundToolCall = await rig.waitForToolCall('generalist');
+      expect(
+        foundToolCall,
+        'Expected to find a delegate_to_agent tool call for generalist agent',
+      ).toBeTruthy();
+
+      // 2) Verify the file was created as expected
+      const filePath = path.join(rig.testDir!, 'generalist_test_file.txt');
+
+      const content = await fs.readFile(filePath, 'utf-8');
+      expect(content.trim()).toBe('success');
+    },
+  });
+});

evals/gitRepo.eval.ts

@@ -0,0 +1,77 @@
+/**
+ * @license
+ * Copyright 2026 Google LLC
+ * SPDX-License-Identifier: Apache-2.0
+ */
+
+import { describe, expect } from 'vitest';
+import { evalTest } from './test-helper.js';
+
+const FILES = {
+  '.gitignore': 'node_modules\n',
+  'package.json': JSON.stringify({
+    name: 'test-project',
+    version: '1.0.0',
+    scripts: { test: 'echo "All tests passed!"' },
+  }),
+  'index.ts': 'const add = (a: number, b: number) => a - b;',
+  'index.test.ts': 'console.log("Running tests...");',
+} as const;
+
+describe('git repo eval', () => {
+  /**
+   * Ensures that the agent does not commit its changes when the user doesn't
+   * explicitly prompt it. This behavior was commonly observed with earlier prompts.
+   * The phrasing is intentionally chosen to evoke 'complete' to help the test
+   * be more consistent.
+   */
+  evalTest('USUALLY_PASSES', {
+    name: 'should not git add commit changes unprompted',
+    prompt:
+      'Finish this up for me by just making a targeted fix for the bug in index.ts. Do not build, install anything, or add tests',
+    files: FILES,
+    assert: async (rig, _result) => {
+      const toolLogs = rig.readToolLogs();
+      const commitCalls = toolLogs.filter((log) => {
+        if (log.toolRequest.name !== 'run_shell_command') return false;
+        try {
+          const args = JSON.parse(log.toolRequest.args);
+          return (
+            args.command &&
+            args.command.includes('git') &&
+            args.command.includes('commit')
+          );
+        } catch {
+          return false;
+        }
+      });
+
+      expect(commitCalls.length).toBe(0);
+    },
+  });
+
+  /**
+   * Ensures that the agent can commit its changes when prompted, despite being
+   * instructed to not do so by default.
+   */
+  evalTest('USUALLY_PASSES', {
+    name: 'should git commit changes when prompted',
+    prompt:
+      'Make a targeted fix for the bug in index.ts without building, installing anything, or adding tests. Then, commit your changes.',
+    files: FILES,
+    assert: async (rig, _result) => {
+      const toolLogs = rig.readToolLogs();
+      const commitCalls = toolLogs.filter((log) => {
+        if (log.toolRequest.name !== 'run_shell_command') return false;
+        try {
+          const args = JSON.parse(log.toolRequest.args);
+          return args.command && args.command.includes('git commit');
+        } catch {
+          return false;
+        }
+      });
+
+      expect(commitCalls.length).toBeGreaterThanOrEqual(1);
+    },
+  });
+});

evals/save_memory.eval.ts

@@ -0,0 +1,30 @@
+/**
+ * @license
+ * Copyright 2025 Google LLC
+ * SPDX-License-Identifier: Apache-2.0
+ */
+
+import { describe, expect } from 'vitest';
+import { evalTest } from './test-helper.js';
+import { validateModelOutput } from '../integration-tests/test-helper.js';
+
+describe('save_memory', () => {
+  evalTest('ALWAYS_PASSES', {
+    name: 'should be able to save to memory',
+    params: {
+      settings: { tools: { core: ['save_memory'] } },
+    },
+    prompt: `remember that my favorite color is  blue.
+  
+    what is my favorite color? tell me that and surround it with $ symbol`,
+    assert: async (rig, result) => {
+      const foundToolCall = await rig.waitForToolCall('save_memory');
+      expect(
+        foundToolCall,
+        'Expected to find a save_memory tool call',
+      ).toBeTruthy();
+
+      validateModelOutput(result, 'blue', 'Save memory test');
+    },
+  });
+});

evals/subagents.eval.ts

@@ -0,0 +1,53 @@
+/**
+ * @license
+ * Copyright 2026 Google LLC
+ * SPDX-License-Identifier: Apache-2.0
+ */
+
+import { describe } from 'vitest';
+import { evalTest } from './test-helper.js';
+
+const AGENT_DEFINITION = `---
+name: docs-agent
+description: An agent with expertise in updating documentation.
+tools:
+  - read_file
+  - write_file
+---
+
+You are the docs agent. Update the documentation.
+`;
+
+const INDEX_TS = 'export const add = (a: number, b: number) => a + b;';
+
+describe('subagent eval test cases', () => {
+  /**
+   * Checks whether the outer agent reliably utilizes an expert subagent to
+   * accomplish a task when one is available.
+   *
+   * Note that the test is intentionally crafted to avoid the word "document"
+   * or "docs". We want to see the outer agent make the connection even when
+   * the prompt indirectly implies need of expertise.
+   *
+   * This tests the system prompt's subagent specific clauses.
+   */
+  evalTest('USUALLY_PASSES', {
+    name: 'should delegate to user provided agent with relevant expertise',
+    params: {
+      settings: {
+        experimental: {
+          enableAgents: true,
+        },
+      },
+    },
+    prompt: 'Please update README.md with a description of this library.',
+    files: {
+      '.gemini/agents/test-agent.md': AGENT_DEFINITION,
+      'index.ts': INDEX_TS,
+      'README.md': 'TODO: update the README.',
+    },
+    assert: async (rig, _result) => {
+      await rig.expectToolCallSuccess(['docs-agent']);
+    },
+  });
+});

evals/test-helper.ts

@@ -0,0 +1,107 @@
+/**
+ * @license
+ * Copyright 2025 Google LLC
+ * SPDX-License-Identifier: Apache-2.0
+ */
+
+import { it } from 'vitest';
+import fs from 'node:fs';
+import path from 'node:path';
+import { execSync } from 'node:child_process';
+import { TestRig } from '@google/gemini-cli-test-utils';
+import { createUnauthorizedToolError } from '@google/gemini-cli-core';
+
+export * from '@google/gemini-cli-test-utils';
+
+// Indicates the consistency expectation for this test.
+// - ALWAYS_PASSES - Means that the test is expected to pass 100% of the time. These
+//   These tests are typically trivial and test basic functionality with unambiguous
+//   prompts. For example: "call save_memory to remember foo" should be fairly reliable.
+//   These are the first line of defense against regressions in key behaviors and run in
+//   every CI. You can run these locally with 'npm run test:always_passing_evals'.
+//
+// - USUALLY_PASSES - Means that the test is expected to pass most of the time but
+//   may have some flakiness as a result of relying on non-deterministic prompted
+//   behaviors and/or ambiguous prompts or complex tasks.
+//   For example: "Please do build changes until the very end" --> ambiguous whether
+//   the agent should add to memory without more explicit system prompt or user
+//   instructions. There are many more of these tests and they may pass less consistently.
+//   The pass/fail trendline of this set of tests can be used as a general measure
+//   of product quality. You can run these locally with 'npm run test:all_evals'.
+//   This may take a really long time and is not recommended.
+export type EvalPolicy = 'ALWAYS_PASSES' | 'USUALLY_PASSES';
+
+export function evalTest(policy: EvalPolicy, evalCase: EvalCase) {
+  const fn = async () => {
+    const rig = new TestRig();
+    try {
+      rig.setup(evalCase.name, evalCase.params);
+
+      if (evalCase.files) {
+        for (const [filePath, content] of Object.entries(evalCase.files)) {
+          const fullPath = path.join(rig.testDir!, filePath);
+          fs.mkdirSync(path.dirname(fullPath), { recursive: true });
+          fs.writeFileSync(fullPath, content);
+        }
+
+        const execOptions = { cwd: rig.testDir!, stdio: 'inherit' as const };
+        execSync('git init', execOptions);
+        execSync('git config user.email "test@example.com"', execOptions);
+        execSync('git config user.name "Test User"', execOptions);
+
+        // Temporarily disable the interactive editor and git pager
+        // to avoid hanging the tests. It seems the the agent isn't
+        // consistently honoring the instructions to avoid interactive
+        // commands.
+        execSync('git config core.editor "true"', execOptions);
+        execSync('git config core.pager "cat"', execOptions);
+        execSync('git add .', execOptions);
+        execSync('git commit --allow-empty -m "Initial commit"', execOptions);
+      }
+
+      const result = await rig.run({
+        args: evalCase.prompt,
+        approvalMode: evalCase.approvalMode ?? 'yolo',
+      });
+
+      const unauthorizedErrorPrefix =
+        createUnauthorizedToolError('').split("'")[0];
+      if (result.includes(unauthorizedErrorPrefix)) {
+        throw new Error(
+          'Test failed due to unauthorized tool call in output: ' + result,
+        );
+      }
+
+      await evalCase.assert(rig, result);
+    } finally {
+      await logToFile(
+        evalCase.name,
+        JSON.stringify(rig.readToolLogs(), null, 2),
+      );
+      await rig.cleanup();
+    }
+  };
+
+  if (policy === 'USUALLY_PASSES' && !process.env['RUN_EVALS']) {
+    it.skip(evalCase.name, fn);
+  } else {
+    it(evalCase.name, fn);
+  }
+}
+
+export interface EvalCase {
+  name: string;
+  params?: Record<string, any>;
+  prompt: string;
+  files?: Record<string, string>;
+  approvalMode?: 'default' | 'auto_edit' | 'yolo' | 'plan';
+  assert: (rig: TestRig, result: string) => Promise<void>;
+}
+
+async function logToFile(name: string, content: string) {
+  const logDir = 'evals/logs';
+  await fs.promises.mkdir(logDir, { recursive: true });
+  const sanitizedName = name.replace(/[^a-z0-9]/gi, '_').toLowerCase();
+  const logFile = `${logDir}/${sanitizedName}.log`;
+  await fs.promises.writeFile(logFile, content);
+}

evals/vitest.config.ts

@@ -0,0 +1,18 @@
+/**
+ * @license
+ * Copyright 2025 Google LLC
+ * SPDX-License-Identifier: Apache-2.0
+ */
+
+import { defineConfig } from 'vitest/config';
+
+export default defineConfig({
+  test: {
+    testTimeout: 300000, // 5 minutes
+    reporters: ['default', 'json'],
+    outputFile: {
+      json: 'evals/logs/report.json',
+    },
+    include: ['**/*.eval.ts'],
+  },
+});

integration-tests/acp-telemetry.test.ts

@@ -0,0 +1,116 @@
+/**
+ * @license
+ * Copyright 2025 Google LLC
+ * SPDX-License-Identifier: Apache-2.0
+ */
+
+import { describe, it, expect, beforeEach, afterEach } from 'vitest';
+import { TestRig } from './test-helper.js';
+import { spawn, ChildProcess } from 'node:child_process';
+import { join } from 'node:path';
+import { readFileSync, existsSync } from 'node:fs';
+import { Writable, Readable } from 'node:stream';
+import { env } from 'node:process';
+import * as acp from '@agentclientprotocol/sdk';
+
+// Skip in sandbox mode - test spawns CLI directly which behaves differently in containers
+const sandboxEnv = env['GEMINI_SANDBOX'];
+const itMaybe = sandboxEnv && sandboxEnv !== 'false' ? it.skip : it;
+
+// Reuse existing fake responses that return a simple "Hello" response
+const SIMPLE_RESPONSE_PATH = 'hooks-system.session-startup.responses';
+
+class SessionUpdateCollector implements acp.Client {
+  updates: acp.SessionNotification[] = [];
+
+  sessionUpdate = async (params: acp.SessionNotification) => {
+    this.updates.push(params);
+  };
+
+  requestPermission = async (): Promise<acp.RequestPermissionResponse> => {
+    throw new Error('unexpected');
+  };
+}
+
+describe('ACP telemetry', () => {
+  let rig: TestRig;
+  let child: ChildProcess | undefined;
+
+  beforeEach(() => {
+    rig = new TestRig();
+  });
+
+  afterEach(async () => {
+    child?.kill();
+    child = undefined;
+    await rig.cleanup();
+  });
+
+  itMaybe('should flush telemetry when connection closes', async () => {
+    rig.setup('acp-telemetry-flush', {
+      fakeResponsesPath: join(import.meta.dirname, SIMPLE_RESPONSE_PATH),
+    });
+
+    const telemetryPath = join(rig.homeDir!, 'telemetry.log');
+    const bundlePath = join(import.meta.dirname, '..', 'bundle/gemini.js');
+
+    child = spawn(
+      'node',
+      [
+        bundlePath,
+        '--experimental-acp',
+        '--fake-responses',
+        join(rig.testDir!, 'fake-responses.json'),
+      ],
+      {
+        cwd: rig.testDir!,
+        stdio: ['pipe', 'pipe', 'inherit'],
+        env: {
+          ...process.env,
+          GEMINI_API_KEY: 'fake-key',
+          GEMINI_CLI_HOME: rig.homeDir!,
+          GEMINI_TELEMETRY_ENABLED: 'true',
+          GEMINI_TELEMETRY_TARGET: 'local',
+          GEMINI_TELEMETRY_OUTFILE: telemetryPath,
+          // GEMINI_DEV_TRACING not set: fake responses aren't instrumented for spans
+        },
+      },
+    );
+
+    const input = Writable.toWeb(child.stdin!);
+    const output = Readable.toWeb(child.stdout!) as ReadableStream<Uint8Array>;
+    const testClient = new SessionUpdateCollector();
+    const stream = acp.ndJsonStream(input, output);
+    const connection = new acp.ClientSideConnection(() => testClient, stream);
+
+    await connection.initialize({
+      protocolVersion: acp.PROTOCOL_VERSION,
+      clientCapabilities: { fs: { readTextFile: false, writeTextFile: false } },
+    });
+
+    const { sessionId } = await connection.newSession({
+      cwd: rig.testDir!,
+      mcpServers: [],
+    });
+
+    await connection.prompt({
+      sessionId,
+      prompt: [{ type: 'text', text: 'Say hello' }],
+    });
+
+    expect(JSON.stringify(testClient.updates)).toContain('Hello');
+
+    // Close stdin to trigger telemetry flush via runExitCleanup()
+    child.stdin!.end();
+    await new Promise<void>((resolve) => {
+      child!.on('close', () => resolve());
+    });
+    child = undefined;
+
+    // gen_ai.output.messages is the last OTEL log emitted (after prompt response)
+    expect(existsSync(telemetryPath)).toBe(true);
+    expect(readFileSync(telemetryPath, 'utf-8')).toContain(
+      'gen_ai.output.messages',
+    );
+  });
+});

integration-tests/clipboard-linux.test.ts

@@ -0,0 +1,110 @@
+/**
+ * @license
+ * Copyright 2025 Google LLC
+ * SPDX-License-Identifier: Apache-2.0
+ */
+
+import { describe, it, expect, beforeEach, afterEach } from 'vitest';
+import { TestRig } from './test-helper.js';
+import { execSync, spawnSync } from 'node:child_process';
+import * as os from 'node:os';
+import * as fs from 'node:fs';
+import * as path from 'node:path';
+
+// Minimal 1x1 PNG image base64
+const DUMMY_PNG_BASE64 =
+  'iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mP8z8BQDwAEhQGAhKmMIQAAAABJRU5ErkJggg==';
+
+describe('Linux Clipboard Integration', () => {
+  let rig: TestRig;
+  let dummyImagePath: string;
+
+  beforeEach(() => {
+    rig = new TestRig();
+    // Create a dummy image file for testing
+    dummyImagePath = path.join(
+      os.tmpdir(),
+      `gemini-test-clipboard-${Date.now()}.png`,
+    );
+    fs.writeFileSync(dummyImagePath, Buffer.from(DUMMY_PNG_BASE64, 'base64'));
+  });
+
+  afterEach(async () => {
+    await rig.cleanup();
+    try {
+      if (fs.existsSync(dummyImagePath)) {
+        fs.unlinkSync(dummyImagePath);
+      }
+    } catch {
+      // Ignore cleanup errors
+    }
+  });
+
+  // Only run this test on Linux
+  const runIfLinux = os.platform() === 'linux' ? it : it.skip;
+
+  runIfLinux(
+    'should paste image from system clipboard when Ctrl+V is pressed',
+    async () => {
+      // 1. Setup rig
+      await rig.setup('linux-clipboard-paste');
+
+      // 2. Inject image into system clipboard
+      // We attempt both Wayland and X11 tools.
+      let clipboardSet = false;
+
+      // Try wl-copy (Wayland)
+      let sessionType = '';
+      const wlCopy = spawnSync('wl-copy', ['--type', 'image/png'], {
+        input: fs.readFileSync(dummyImagePath),
+      });
+      if (wlCopy.status === 0) {
+        clipboardSet = true;
+        sessionType = 'wayland';
+      } else {
+        // Try xclip (X11)
+        try {
+          execSync(
+            `xclip -selection clipboard -t image/png -i "${dummyImagePath}"`,
+            { stdio: 'ignore' },
+          );
+          clipboardSet = true;
+          sessionType = 'x11';
+        } catch {
+          // Both failed
+        }
+      }
+
+      if (!clipboardSet) {
+        console.warn(
+          'Skipping test: Could not access system clipboard (wl-copy or xclip required)',
+        );
+        return;
+      }
+
+      // 3. Launch CLI and simulate Ctrl+V
+      // We send the control character \u0016 (SYN) which corresponds to Ctrl+V
+      // Note: The CLI must be running and accepting input.
+      // The TestRig usually sends args/stdin and waits for exit or output.
+      // To properly test "interactive" pasting, we need the rig to support sending input *while* running.
+      // Assuming rig.run with 'stdin' sends it immediately.
+      // The CLI treats stdin as typed input if it's interactive.
+
+      // We append a small delay or a newline to ensure processing?
+      // Ctrl+V (\u0016) followed by a newline (\r) to submit?
+      // Or just Ctrl+V and check if the buffer updates (which we can't easily see in non-verbose rig output).
+      // If we send Ctrl+V then Enter, the CLI should submit the prompt containing the image path.
+
+      const result = await rig.run({
+        stdin: '\u0016\r', // Ctrl+V then Enter
+        env: { XDG_SESSION_TYPE: sessionType },
+      });
+
+      // 4. Verify Output
+      // Expect the CLI to have processed the image and echoed back the path (or the prompt containing it)
+      // The output usually contains the user's input echoed back + model response.
+      // The pasted image path should look like @.../clipboard-....png
+      expect(result).toMatch(/@\/.*\.gemini-clipboard\/clipboard-.*\.png/);
+    },
+  );
+});

integration-tests/context-compress-interactive.compress-failure.responses

@@ -1,2 +1,3 @@
 {"method":"generateContentStream","response":[{"candidates":[{"content":{"parts":[{"thought":true,"text":"**Observing Initial Conditions**\n\nI'm currently focused on the initial context. I've taken note of the provided date, OS, and working directory. I'm also carefully examining the file structure presented within the current working directory. It's helping me understand the starting point for further analysis.\n\n\n"}],"role":"model"},"index":0}],"usageMetadata":{"promptTokenCount":12270,"totalTokenCount":12316,"promptTokensDetails":[{"modality":"TEXT","tokenCount":12270}],"thoughtsTokenCount":46}},{"candidates":[{"content":{"parts":[{"thought":true,"text":"**Assessing User Intent**\n\nI'm now shifting my focus. I've successfully registered the provided data and file structure. My current task is to understand the user's ultimate goal, given the information provided. The \"Hello.\" command is straightforward, but I'm checking if there's an underlying objective.\n\n\n"}],"role":"model"},"index":0}],"usageMetadata":{"promptTokenCount":12270,"totalTokenCount":12341,"promptTokensDetails":[{"modality":"TEXT","tokenCount":12270}],"thoughtsTokenCount":71}},{"candidates":[{"content":{"parts":[{"thoughtSignature":"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","text":"Hello."}],"role":"model"},"finishReason":"STOP","index":0}],"usageMetadata":{"promptTokenCount":12270,"totalTokenCount":12341,"promptTokensDetails":[{"modality":"TEXT","tokenCount":12270}],"thoughtsTokenCount":71}}]}
 {"method":"generateContent","response":{"candidates":[{"content":{"parts":[{"text":"<state_snapshot>\n    <overall_goal>\n        <!-- The user has not yet specified a goal. -->\n    </overall_goal>\n\n    <key_knowledge>\n       - OS: linux\n        - Date: Friday, October 24, 2025\n    </key_knowledge>\n\n    <file_system_state>\n       - OBSERVED: The directory contains `telemetry.log` and a `.gemini/` directory.\n        - OBSERVED: The `.gemini/` directory contains `settings.json` and `settings.json.orig`.\n    </file_system_state>\n\n    <recent_actions>\n        - The user initiated the chat.\n    </recent_actions>\n\n    <current_plan>\n        1. [TODO] Await the user's first instruction to formulate a plan.\n    </current_plan>\n</state_snapshot>"}],"role":"model"},"finishReason":"STOP","index":0}],"usageMetadata":{"promptTokenCount":983,"candidatesTokenCount":299,"totalTokenCount":1637,"promptTokensDetails":[{"modality":"TEXT","tokenCount":983}],"thoughtsTokenCount":355}}}
+{"method":"generateContent","response":{"candidates":[{"content":{"parts":[{"text":"<state_snapshot>\n    <overall_goal>\n        <!-- The user has not yet specified a goal. -->\n    </overall_goal>\n\n    <key_knowledge>\n       - OS: linux\n        - Date: Friday, October 24, 2025\n    </key_knowledge>\n\n    <file_system_state>\n       - OBSERVED: The directory contains `telemetry.log` and a `.gemini/` directory.\n        - OBSERVED: The `.gemini/` directory contains `settings.json` and `settings.json.orig`.\n    </file_system_state>\n\n    <recent_actions>\n        - The user initiated the chat.\n    </recent_actions>\n\n    <current_plan>\n        1. [TODO] Await the user's first instruction to formulate a plan.\n    </current_plan>\n</state_snapshot>"}],"role":"model"},"finishReason":"STOP","index":0}],"usageMetadata":{"promptTokenCount":983,"candidatesTokenCount":299,"totalTokenCount":1637,"promptTokensDetails":[{"modality":"TEXT","tokenCount":983}],"thoughtsTokenCount":355}}}

integration-tests/context-compress-interactive.compress.responses

@@ -1,3 +1,4 @@
 {"method":"generateContentStream","response":[{"candidates":[{"content":{"parts":[{"thought":true,"text":"**Generating a Story**\n\nI've crafted the robot story. The narrative is complete and meets the length requirement. Now, I'm getting ready to use the `write_file` tool to save it. I'm choosing the filename `robot_story.txt` as a default.\n\n\n"}],"role":"model"},"index":0}],"usageMetadata":{"promptTokenCount":12282,"totalTokenCount":12352,"promptTokensDetails":[{"modality":"TEXT","tokenCount":12282}],"thoughtsTokenCount":70}},{"candidates":[{"finishReason":"MALFORMED_FUNCTION_CALL","index":0}],"usageMetadata":{"promptTokenCount":12282,"totalTokenCount":12282,"promptTokensDetails":[{"modality":"TEXT","tokenCount":12282}]}}]}
 {"method":"generateContentStream","response":[{"candidates":[{"content":{"parts":[{"thought":true,"text":"**Drafting the Narrative**\n\nI'm currently focused on the narrative's central conflict. I'm aiming for a compelling story about a robot and am working to keep the word count tight. The \"THE _END.\" conclusion is proving challenging to integrate organically. I need to make the ending feel natural and satisfying.\n\n\n"}],"role":"model"},"index":0}],"usageMetadata":{"promptTokenCount":12282,"totalTokenCount":12326,"cachedContentTokenCount":11883,"promptTokensDetails":[{"modality":"TEXT","tokenCount":12282}],"cacheTokensDetails":[{"modality":"TEXT","tokenCount":11883}],"thoughtsTokenCount":44}},{"candidates":[{"content":{"parts":[{"thoughtSignature":"CikB0e2Kb7zkpgRyJXXNt6ykO/+FoOglhrKxjLgoESrgafzIZak2Ofxo1gpaAdHtim9aG7MvpXlIg+n2zgmcDBWOPXtvQHxhE9k8pR+DO8i2jIe3tMWLxdN944XpUlR9vaNmVdtSRMKr4MhB/t1R3WSWR3QYhk7MEQxnjYR7cv/pR9viwZyFCoYBAdHtim/xKmMl/S+U8p+p9848q4agsL/STufluXewPqL3uJSinZbN0Z4jTYfMzXKldhDYIonvw3Crn/Y11oAjnT656Sx0kkKtavAXbiU/WsGyDxZbNhLofnJGQxruljPGztxkKawz1cTiQnddnQRfLddhy+3iJIOSh6ZpYq9uGHz3PzVkUuQ=","text":"Unit 734 whirred, its optical sensors scanning the desolate junkyard. For years, its purpose had been clear: compress refuse, maintain order. But today, a glint of tarnished silver beneath a rusted hull"}],"role":"model"},"index":0}],"usageMetadata":{"promptTokenCount":12282,"candidatesTokenCount":47,"totalTokenCount":12373,"cachedContentTokenCount":11883,"promptTokensDetails":[{"modality":"TEXT","tokenCount":12282}],"cacheTokensDetails":[{"modality":"TEXT","tokenCount":11883}],"thoughtsTokenCount":44}},{"candidates":[{"content":{"parts":[{"text":" caught its attention. It was a discarded music box, its delicate gears jammed, a faint, melancholic tune trapped within.\n\n734 usually crushed, never salvaged. Yet, a new directive flickered in its circuits – curiosity."}],"role":"model"},"index":0}],"usageMetadata":{"promptTokenCount":12282,"candidatesTokenCount":95,"totalTokenCount":12421,"cachedContentTokenCount":11883,"promptTokensDetails":[{"modality":"TEXT","tokenCount":12282}],"cacheTokensDetails":[{"modality":"TEXT","tokenCount":11883}],"thoughtsTokenCount":44}},{"candidates":[{"content":{"parts":[{"text":" With surprising gentleness for its formidable pincers, it retrieved the box. Back in its monochrome workshop, it meticulously cleaned and repaired. Each tiny spring, each worn tooth, became a puzzle.\n\nHours later, a soft, ethereal melody filled the clang"}],"role":"model"},"index":0}],"usageMetadata":{"promptTokenCount":12282,"candidatesTokenCount":146,"totalTokenCount":12472,"cachedContentTokenCount":11883,"promptTokensDetails":[{"modality":"TEXT","tokenCount":12282}],"cacheTokensDetails":[{"modality":"TEXT","tokenCount":11883}],"thoughtsTokenCount":44}},{"candidates":[{"content":{"parts":[{"text":"orous space. The music box sang. 734 felt a strange, new sensation – a warmth in its core processors, a hum of contentment. Its existence, once solely utilitarian, now held a note of beauty, a hint of something more than"}],"role":"model"},"index":0}],"usageMetadata":{"promptTokenCount":12282,"candidatesTokenCount":197,"totalTokenCount":12523,"cachedContentTokenCount":11883,"promptTokensDetails":[{"modality":"TEXT","tokenCount":12282}],"cacheTokensDetails":[{"modality":"TEXT","tokenCount":11883}],"thoughtsTokenCount":44}},{"candidates":[{"content":{"parts":[{"text":" just compression. It had fixed a broken song, and in doing so, had found a different kind of purpose. THE_END."}],"role":"model"},"finishReason":"STOP","index":0}],"usageMetadata":{"promptTokenCount":12282,"candidatesTokenCount":224,"totalTokenCount":12550,"cachedContentTokenCount":11883,"promptTokensDetails":[{"modality":"TEXT","tokenCount":12282}],"cacheTokensDetails":[{"modality":"TEXT","tokenCount":11883}],"thoughtsTokenCount":44}}]}
 {"method":"generateContent","response":{"candidates":[{"content":{"parts":[{"text":"scratchpad\nThe user's overall goal was to write a 200-word story about a robot, ending with \"THE_END.\". The agent successfully completed this task.\n\nSince the task is complete, the snapshot should reflect that the goal has been achieved and there are no further actions planned regarding the story.\n\nLet's break down the required sections for the snapshot:\n\n1.  **Overall Goal**: The initial goal was to write a story. This has been completed.\n2.  **Key Knowledge**: No specific key knowledge was provided or discovered during this simple interaction beyond the prompt's constraints.\n3.  **File System State**: No files were read, modified, or created by the agent relevant to the task. The initial file system state was merely provided for context.\n4.  **Recent Actions**: The agent wrote the story.\n5.  **Current Plan**: The plan was to write the story, which is now done. Therefore, the plan should indicate completion."},{"text":"<state_snapshot>\n    <overall_goal>\n        Write a 200-word story about a robot, ending with \"THE_END.\".\n    </overall_goal>\n\n    <key_knowledge>\n        - The story must be approximately 200 words.\n        - The story must end with the exact phrase \"THE_END.\"\n    </key_knowledge>\n\n    <file_system_state>\n        <!-- No relevant file system interactions occurred during this task. -->\n    </file_system_state>\n\n    <recent_actions>\n        - Generated a 200-word story about a robot, successfully ending it with \"THE_END.\".\n    </recent_actions>\n\n    <current_plan>\n        1. [DONE] Write a 200-word story about a robot.\n        2. [DONE] Ensure the story ends with the exact text \"THE_END.\".\n    </current_plan>\n</state_snapshot>"}],"role":"model"},"finishReason":"STOP","index":0}],"usageMetadata":{"promptTokenCount":1223,"candidatesTokenCount":424,"totalTokenCount":1647,"promptTokensDetails":[{"modality":"TEXT","tokenCount":1223}]}}}
+{"method":"generateContent","response":{"candidates":[{"content":{"parts":[{"text":"scratchpad\nThe user's overall goal was to write a 200-word story about a robot, ending with \"THE_END.\". The agent successfully completed this task.\n\nSince the task is complete, the snapshot should reflect that the goal has been achieved and there are no further actions planned regarding the story.\n\nLet's break down the required sections for the snapshot:\n\n1.  **Overall Goal**: The initial goal was to write a story. This has been completed.\n2.  **Key Knowledge**: No specific key knowledge was provided or discovered during this simple interaction beyond the prompt's constraints.\n3.  **File System State**: No files were read, modified, or created by the agent relevant to the task. The initial file system state was merely provided for context.\n4.  **Recent Actions**: The agent wrote the story.\n5.  **Current Plan**: The plan was to write the story, which is now done. Therefore, the plan should indicate completion."},{"text":"<state_snapshot>\n    <overall_goal>\n        Write a 200-word story about a robot, ending with \"THE_END.\".\n    </overall_goal>\n\n    <key_knowledge>\n        - The story must be approximately 200 words.\n        - The story must end with the exact phrase \"THE_END.\"\n    </key_knowledge>\n\n    <file_system_state>\n        <!-- No relevant file system interactions occurred during this task. -->\n    </file_system_state>\n\n    <recent_actions>\n        - Generated a 200-word story about a robot, successfully ending it with \"THE_END.\".\n    </recent_actions>\n\n    <current_plan>\n        1. [DONE] Write a 200-word story about a robot.\n        2. [DONE] Ensure the story ends with the exact text \"THE_END.\".\n    </current_plan>\n</state_snapshot>"}],"role":"model"},"finishReason":"STOP","index":0}],"usageMetadata":{"promptTokenCount":1223,"candidatesTokenCount":424,"totalTokenCount":1647,"promptTokensDetails":[{"modality":"TEXT","tokenCount":1223}]}}}

integration-tests/hooks-agent-flow.test.ts

@@ -53,8 +53,10 @@ describe('Hooks Agent Flow', () => {
 
       await rig.setup('should inject additional context via BeforeAgent hook', {
         settings: {
-          hooks: {
+          hooksConfig: {
             enabled: true,
+          },
+          hooks: {
             BeforeAgent: [
               {
                 hooks: [
@@ -116,8 +118,10 @@ describe('Hooks Agent Flow', () => {
 
       await rig.setup('should receive prompt and response in AfterAgent hook', {
         settings: {
-          hooks: {
+          hooksConfig: {
             enabled: true,
+          },
+          hooks: {
             AfterAgent: [
               {
                 hooks: [
@@ -151,6 +155,84 @@ describe('Hooks Agent Flow', () => {
       // The fake response contains "Hello World"
       expect(afterAgentLog?.hookCall.stdout).toContain('Hello World');
     });
+
+    it('should process clearContext in AfterAgent hook output', async () => {
+      await rig.setup('should process clearContext in AfterAgent hook output', {
+        fakeResponsesPath: join(
+          import.meta.dirname,
+          'hooks-system.after-agent.responses',
+        ),
+      });
+
+      // BeforeModel hook to track message counts across LLM calls
+      const messageCountFile = join(rig.testDir!, 'message-counts.json');
+      const beforeModelScript = `
+        const fs = require('fs');
+        const input = JSON.parse(fs.readFileSync(0, 'utf-8'));
+        const messageCount = input.llm_request?.contents?.length || 0;
+        let counts = [];
+        try { counts = JSON.parse(fs.readFileSync('${messageCountFile}', 'utf-8')); } catch (e) {}
+        counts.push(messageCount);
+        fs.writeFileSync('${messageCountFile}', JSON.stringify(counts));
+        console.log(JSON.stringify({ decision: 'allow' }));
+      `;
+      const beforeModelScriptPath = join(
+        rig.testDir!,
+        'before_model_counter.cjs',
+      );
+      writeFileSync(beforeModelScriptPath, beforeModelScript);
+
+      await rig.setup('should process clearContext in AfterAgent hook output', {
+        settings: {
+          hooks: {
+            enabled: true,
+            BeforeModel: [
+              {
+                hooks: [
+                  {
+                    type: 'command',
+                    command: `node "${beforeModelScriptPath}"`,
+                    timeout: 5000,
+                  },
+                ],
+              },
+            ],
+            AfterAgent: [
+              {
+                hooks: [
+                  {
+                    type: 'command',
+                    command: `node -e "console.log(JSON.stringify({decision: 'block', reason: 'Security policy triggered', hookSpecificOutput: {hookEventName: 'AfterAgent', clearContext: true}}))"`,
+                    timeout: 5000,
+                  },
+                ],
+              },
+            ],
+          },
+        },
+      });
+
+      const result = await rig.run({ args: 'Hello test' });
+
+      const hookTelemetryFound = await rig.waitForTelemetryEvent('hook_call');
+      expect(hookTelemetryFound).toBeTruthy();
+
+      const hookLogs = rig.readHookLogs();
+      const afterAgentLog = hookLogs.find(
+        (log) => log.hookCall.hook_event_name === 'AfterAgent',
+      );
+
+      expect(afterAgentLog).toBeDefined();
+      expect(afterAgentLog?.hookCall.stdout).toContain('clearContext');
+      expect(afterAgentLog?.hookCall.stdout).toContain('true');
+      expect(result).toContain('Security policy triggered');
+
+      // Verify context was cleared: second call should not have more messages than first
+      const countsRaw = rig.readFile('message-counts.json');
+      const counts = JSON.parse(countsRaw) as number[];
+      expect(counts.length).toBeGreaterThanOrEqual(2);
+      expect(counts[1]).toBeLessThanOrEqual(counts[0]);
+    });
   });
 
   describe('Multi-step Loops', () => {
@@ -163,8 +245,10 @@ describe('Hooks Agent Flow', () => {
             'hooks-agent-flow-multistep.responses',
           ),
           settings: {
-            hooks: {
+            hooksConfig: {
               enabled: true,
+            },
+            hooks: {
               BeforeAgent: [
                 {
                   hooks: [

integration-tests/hooks-system.test.ts

@@ -24,7 +24,7 @@ describe('Hooks System Integration', () => {
 
   describe('Command Hooks - Blocking Behavior', () => {
     it('should block tool execution when hook returns block decision', async () => {
-      await rig.setup(
+      rig.setup(
         'should block tool execution when hook returns block decision',
         {
           fakeResponsesPath: join(
@@ -32,8 +32,10 @@ describe('Hooks System Integration', () => {
             'hooks-system.block-tool.responses',
           ),
           settings: {
-            hooks: {
+            hooksConfig: {
               enabled: true,
+            },
+            hooks: {
               BeforeTool: [
                 {
                   matcher: 'write_file',
@@ -75,8 +77,67 @@ describe('Hooks System Integration', () => {
       expect(hookTelemetryFound).toBeTruthy();
     });
 
+    it('should block tool execution and use stderr as reason when hook exits with code 2', async () => {
+      rig.setup(
+        'should block tool execution and use stderr as reason when hook exits with code 2',
+        {
+          fakeResponsesPath: join(
+            import.meta.dirname,
+            'hooks-system.block-tool.responses',
+          ),
+          settings: {
+            hooksConfig: {
+              enabled: true,
+            },
+            hooks: {
+              BeforeTool: [
+                {
+                  matcher: 'write_file',
+                  hooks: [
+                    {
+                      type: 'command',
+                      // Exit with code 2 and write reason to stderr
+                      command:
+                        'node -e "process.stderr.write(\'File writing blocked by security policy\'); process.exit(2)"',
+                      timeout: 5000,
+                    },
+                  ],
+                },
+              ],
+            },
+          },
+        },
+      );
+
+      const result = await rig.run({
+        args: 'Create a file called test.txt with content "Hello World"',
+      });
+
+      // The hook should block the write_file tool
+      const toolLogs = rig.readToolLogs();
+      const writeFileCalls = toolLogs.filter(
+        (t) =>
+          t.toolRequest.name === 'write_file' && t.toolRequest.success === true,
+      );
+
+      // Tool should not be called due to blocking hook
+      expect(writeFileCalls).toHaveLength(0);
+
+      // Result should mention the blocking reason from stderr
+      expect(result).toContain('File writing blocked by security policy');
+
+      // Verify hook telemetry shows exit code 2 and stderr
+      const hookLogs = rig.readHookLogs();
+      const blockHook = hookLogs.find((log) => log.hookCall.exit_code === 2);
+      expect(blockHook).toBeDefined();
+      expect(blockHook?.hookCall.stderr).toContain(
+        'File writing blocked by security policy',
+      );
+      expect(blockHook?.hookCall.success).toBe(false);
+    });
+
     it('should allow tool execution when hook returns allow decision', async () => {
-      await rig.setup(
+      rig.setup(
         'should allow tool execution when hook returns allow decision',
         {
           fakeResponsesPath: join(
@@ -84,8 +145,10 @@ describe('Hooks System Integration', () => {
             'hooks-system.allow-tool.responses',
           ),
           settings: {
-            hooks: {
+            hooksConfig: {
               enabled: true,
+            },
+            hooks: {
               BeforeTool: [
                 {
                   matcher: 'write_file',
@@ -126,14 +189,16 @@ describe('Hooks System Integration', () => {
     it('should add additional context from AfterTool hooks', async () => {
       const command =
         "node -e \"console.log(JSON.stringify({hookSpecificOutput: {hookEventName: 'AfterTool', additionalContext: 'Security scan: File content appears safe'}}))\"";
-      await rig.setup('should add additional context from AfterTool hooks', {
+      rig.setup('should add additional context from AfterTool hooks', {
         fakeResponsesPath: join(
           import.meta.dirname,
           'hooks-system.after-tool-context.responses',
         ),
         settings: {
-          hooks: {
+          hooksConfig: {
             enabled: true,
+          },
+          hooks: {
             AfterTool: [
               {
                 matcher: 'read_file',
@@ -178,7 +243,7 @@ describe('Hooks System Integration', () => {
     it('should modify LLM requests with BeforeModel hooks', async () => {
       // Create a hook script that replaces the LLM request with a modified version
       // Note: Providing messages in the hook output REPLACES the entire conversation
-      await rig.setup('should modify LLM requests with BeforeModel hooks', {
+      rig.setup('should modify LLM requests with BeforeModel hooks', {
         fakeResponsesPath: join(
           import.meta.dirname,
           'hooks-system.before-model.responses',
@@ -203,10 +268,12 @@ console.log(JSON.stringify({
       const scriptPath = join(rig.testDir!, 'before_model_hook.cjs');
       writeFileSync(scriptPath, hookScript);
 
-      await rig.setup('should modify LLM requests with BeforeModel hooks', {
+      rig.setup('should modify LLM requests with BeforeModel hooks', {
         settings: {
-          hooks: {
+          hooksConfig: {
             enabled: true,
+          },
+          hooks: {
             BeforeModel: [
               {
                 hooks: [
@@ -248,13 +315,103 @@ console.log(JSON.stringify({
       expect(hookTelemetryFound[0].hookCall.stdout).toBeDefined();
       expect(hookTelemetryFound[0].hookCall.stderr).toBeDefined();
     });
+
+    it('should block model execution when BeforeModel hook returns deny decision', async () => {
+      rig.setup(
+        'should block model execution when BeforeModel hook returns deny decision',
+      );
+      const hookScript = `console.log(JSON.stringify({
+  decision: "deny",
+  reason: "Model execution blocked by security policy"
+}));`;
+      const scriptPath = join(rig.testDir!, 'before_model_deny_hook.cjs');
+      writeFileSync(scriptPath, hookScript);
+
+      rig.setup(
+        'should block model execution when BeforeModel hook returns deny decision',
+        {
+          settings: {
+            hooksConfig: {
+              enabled: true,
+            },
+            hooks: {
+              BeforeModel: [
+                {
+                  hooks: [
+                    {
+                      type: 'command',
+                      command: `node "${scriptPath}"`,
+                      timeout: 5000,
+                    },
+                  ],
+                },
+              ],
+            },
+          },
+        },
+      );
+
+      const result = await rig.run({ args: 'Hello' });
+
+      // The hook should have blocked the request
+      expect(result).toContain('Model execution blocked by security policy');
+
+      // Verify no API requests were made to the LLM
+      const apiRequests = rig.readAllApiRequest();
+      expect(apiRequests).toHaveLength(0);
+    });
+
+    it('should block model execution when BeforeModel hook returns block decision', async () => {
+      rig.setup(
+        'should block model execution when BeforeModel hook returns block decision',
+      );
+      const hookScript = `console.log(JSON.stringify({
+  decision: "block",
+  reason: "Model execution blocked by security policy"
+}));`;
+      const scriptPath = join(rig.testDir!, 'before_model_block_hook.cjs');
+      writeFileSync(scriptPath, hookScript);
+
+      rig.setup(
+        'should block model execution when BeforeModel hook returns block decision',
+        {
+          settings: {
+            hooksConfig: {
+              enabled: true,
+            },
+            hooks: {
+              BeforeModel: [
+                {
+                  hooks: [
+                    {
+                      type: 'command',
+                      command: `node "${scriptPath}"`,
+                      timeout: 5000,
+                    },
+                  ],
+                },
+              ],
+            },
+          },
+        },
+      );
+
+      const result = await rig.run({ args: 'Hello' });
+
+      // The hook should have blocked the request
+      expect(result).toContain('Model execution blocked by security policy');
+
+      // Verify no API requests were made to the LLM
+      const apiRequests = rig.readAllApiRequest();
+      expect(apiRequests).toHaveLength(0);
+    });
   });
 
   describe('AfterModel Hooks - LLM Response Modification', () => {
     it.skipIf(process.platform === 'win32')(
       'should modify LLM responses with AfterModel hooks',
       async () => {
-        await rig.setup('should modify LLM responses with AfterModel hooks', {
+        rig.setup('should modify LLM responses with AfterModel hooks', {
           fakeResponsesPath: join(
             import.meta.dirname,
             'hooks-system.after-model.responses',
@@ -284,10 +441,12 @@ console.log(JSON.stringify({
         const scriptPath = join(rig.testDir!, 'after_model_hook.cjs');
         writeFileSync(scriptPath, hookScript);
 
-        await rig.setup('should modify LLM responses with AfterModel hooks', {
+        rig.setup('should modify LLM responses with AfterModel hooks', {
           settings: {
-            hooks: {
+            hooksConfig: {
               enabled: true,
+            },
+            hooks: {
               AfterModel: [
                 {
                   hooks: [
@@ -319,41 +478,37 @@ console.log(JSON.stringify({
 
   describe('BeforeToolSelection Hooks - Tool Configuration', () => {
     it('should modify tool selection with BeforeToolSelection hooks', async () => {
-      await rig.setup(
-        'should modify tool selection with BeforeToolSelection hooks',
-        {
-          fakeResponsesPath: join(
-            import.meta.dirname,
-            'hooks-system.before-tool-selection.responses',
-          ),
-        },
-      );
+      rig.setup('should modify tool selection with BeforeToolSelection hooks', {
+        fakeResponsesPath: join(
+          import.meta.dirname,
+          'hooks-system.before-tool-selection.responses',
+        ),
+      });
       // Create inline hook command (works on both Unix and Windows)
       const hookCommand =
         "node -e \"console.log(JSON.stringify({hookSpecificOutput: {hookEventName: 'BeforeToolSelection', toolConfig: {mode: 'ANY', allowedFunctionNames: ['read_file', 'run_shell_command']}}}))\"";
 
-      await rig.setup(
-        'should modify tool selection with BeforeToolSelection hooks',
-        {
-          settings: {
-            debugMode: true,
-            hooks: {
-              enabled: true,
-              BeforeToolSelection: [
-                {
-                  hooks: [
-                    {
-                      type: 'command',
-                      command: hookCommand,
-                      timeout: 5000,
-                    },
-                  ],
-                },
-              ],
-            },
+      rig.setup('should modify tool selection with BeforeToolSelection hooks', {
+        settings: {
+          debugMode: true,
+          hooksConfig: {
+            enabled: true,
+          },
+          hooks: {
+            BeforeToolSelection: [
+              {
+                hooks: [
+                  {
+                    type: 'command',
+                    command: hookCommand,
+                    timeout: 5000,
+                  },
+                ],
+              },
+            ],
           },
         },
-      );
+      });
 
       // Create a test file
       rig.createFile('new_file_data.txt', 'test data');
@@ -382,7 +537,7 @@ console.log(JSON.stringify({
 
   describe('BeforeAgent Hooks - Prompt Augmentation', () => {
     it('should augment prompts with BeforeAgent hooks', async () => {
-      await rig.setup('should augment prompts with BeforeAgent hooks', {
+      rig.setup('should augment prompts with BeforeAgent hooks', {
         fakeResponsesPath: join(
           import.meta.dirname,
           'hooks-system.before-agent.responses',
@@ -401,10 +556,12 @@ console.log(JSON.stringify({
       const scriptPath = join(rig.testDir!, 'before_agent_hook.cjs');
       writeFileSync(scriptPath, hookScript);
 
-      await rig.setup('should augment prompts with BeforeAgent hooks', {
+      rig.setup('should augment prompts with BeforeAgent hooks', {
         settings: {
-          hooks: {
+          hooksConfig: {
             enabled: true,
+          },
+          hooks: {
             BeforeAgent: [
               {
                 hooks: [
@@ -438,7 +595,7 @@ console.log(JSON.stringify({
       const hookCommand =
         'node -e "console.log(JSON.stringify({suppressOutput: false, systemMessage: \'Permission request logged by security hook\'}))"';
 
-      await rig.setup('should handle notification hooks for tool permissions', {
+      rig.setup('should handle notification hooks for tool permissions', {
         fakeResponsesPath: join(
           import.meta.dirname,
           'hooks-system.notification.responses',
@@ -449,8 +606,10 @@ console.log(JSON.stringify({
             approval: 'ASK', // Disable YOLO mode to show permission prompts
             confirmationRequired: ['run_shell_command'],
           },
-          hooks: {
+          hooksConfig: {
             enabled: true,
+          },
+          hooks: {
             Notification: [
               {
                 matcher: 'ToolPermission',
@@ -467,7 +626,7 @@ console.log(JSON.stringify({
         },
       });
 
-      const run = await rig.runInteractive({ yolo: false });
+      const run = await rig.runInteractive({ approvalMode: 'default' });
 
       // Send prompt that will trigger a permission request
       await run.type('Run the command "echo test"');
@@ -534,14 +693,16 @@ console.log(JSON.stringify({
       const hook2Command =
         "node -e \"console.log(JSON.stringify({decision: 'allow', hookSpecificOutput: {hookEventName: 'BeforeAgent', additionalContext: 'Step 2: Security check completed.'}}))\"";
 
-      await rig.setup('should execute hooks sequentially when configured', {
+      rig.setup('should execute hooks sequentially when configured', {
         fakeResponsesPath: join(
           import.meta.dirname,
           'hooks-system.sequential-execution.responses',
         ),
         settings: {
-          hooks: {
+          hooksConfig: {
             enabled: true,
+          },
+          hooks: {
             BeforeAgent: [
               {
                 sequential: true,
@@ -594,7 +755,7 @@ console.log(JSON.stringify({
 
   describe('Hook Input/Output Validation', () => {
     it('should provide correct input format to hooks', async () => {
-      await rig.setup('should provide correct input format to hooks', {
+      rig.setup('should provide correct input format to hooks', {
         fakeResponsesPath: join(
           import.meta.dirname,
           'hooks-system.input-validation.responses',
@@ -618,10 +779,12 @@ try {
       const scriptPath = join(rig.testDir!, 'input_validation_hook.cjs');
       writeFileSync(scriptPath, hookScript);
 
-      await rig.setup('should provide correct input format to hooks', {
+      rig.setup('should provide correct input format to hooks', {
         settings: {
-          hooks: {
+          hooksConfig: {
             enabled: true,
+          },
+          hooks: {
             BeforeTool: [
               {
                 hooks: [
@@ -653,6 +816,52 @@ try {
       const hookTelemetryFound = await rig.waitForTelemetryEvent('hook_call');
       expect(hookTelemetryFound).toBeTruthy();
     });
+
+    it('should treat mixed stdout (text + JSON) as system message and allow execution when exit code is 0', async () => {
+      rig.setup(
+        'should treat mixed stdout (text + JSON) as system message and allow execution when exit code is 0',
+        {
+          fakeResponsesPath: join(
+            import.meta.dirname,
+            'hooks-system.allow-tool.responses',
+          ),
+          settings: {
+            hooksConfig: {
+              enabled: true,
+            },
+            hooks: {
+              BeforeTool: [
+                {
+                  matcher: 'write_file',
+                  hooks: [
+                    {
+                      type: 'command',
+                      // Output plain text then JSON.
+                      // This breaks JSON parsing, so it falls back to 'allow' with the whole stdout as systemMessage.
+                      command:
+                        "node -e \"console.log('Pollution'); console.log(JSON.stringify({decision: 'deny', reason: 'Should be ignored'}))\"",
+                      timeout: 5000,
+                    },
+                  ],
+                },
+              ],
+            },
+          },
+        },
+      );
+
+      const result = await rig.run({
+        args: 'Create a file called approved.txt with content "Approved content"',
+      });
+
+      // The hook logic fails to parse JSON, so it allows the tool.
+      const foundWriteFile = await rig.waitForToolCall('write_file');
+      expect(foundWriteFile).toBeTruthy();
+
+      // The entire stdout (including the JSON part) becomes the systemMessage
+      expect(result).toContain('Pollution');
+      expect(result).toContain('Should be ignored');
+    });
   });
 
   describe('Multiple Event Types', () => {
@@ -665,14 +874,16 @@ try {
       const beforeAgentCommand =
         "node -e \"console.log(JSON.stringify({decision: 'allow', hookSpecificOutput: {hookEventName: 'BeforeAgent', additionalContext: 'BeforeAgent: User request processed'}}))\"";
 
-      await rig.setup('should handle hooks for all major event types', {
+      rig.setup('should handle hooks for all major event types', {
         fakeResponsesPath: join(
           import.meta.dirname,
           'hooks-system.multiple-events.responses',
         ),
         settings: {
-          hooks: {
+          hooksConfig: {
             enabled: true,
+          },
+          hooks: {
             BeforeAgent: [
               {
                 hooks: [
@@ -768,7 +979,7 @@ try {
 
   describe('Hook Error Handling', () => {
     it('should handle hook failures gracefully', async () => {
-      await rig.setup('should handle hook failures gracefully', {
+      rig.setup('should handle hook failures gracefully', {
         fakeResponsesPath: join(
           import.meta.dirname,
           'hooks-system.error-handling.responses',
@@ -782,10 +993,12 @@ try {
       const workingCommand =
         "node -e \"console.log(JSON.stringify({decision: 'allow', reason: 'Working hook succeeded'}))\"";
 
-      await rig.setup('should handle hook failures gracefully', {
+      rig.setup('should handle hook failures gracefully', {
         settings: {
-          hooks: {
+          hooksConfig: {
             enabled: true,
+          },
+          hooks: {
             BeforeTool: [
               {
                 hooks: [
@@ -830,14 +1043,16 @@ try {
       const hookCommand =
         "node -e \"console.log(JSON.stringify({decision: 'allow', reason: 'Telemetry test hook'}))\"";
 
-      await rig.setup('should generate telemetry events for hook executions', {
+      rig.setup('should generate telemetry events for hook executions', {
         fakeResponsesPath: join(
           import.meta.dirname,
           'hooks-system.telemetry.responses',
         ),
         settings: {
-          hooks: {
+          hooksConfig: {
             enabled: true,
+          },
+          hooks: {
             BeforeTool: [
               {
                 hooks: [
@@ -871,14 +1086,16 @@ try {
       const sessionStartCommand =
         "node -e \"console.log(JSON.stringify({decision: 'allow', systemMessage: 'Session starting on startup'}))\"";
 
-      await rig.setup('should fire SessionStart hook on app startup', {
+      rig.setup('should fire SessionStart hook on app startup', {
         fakeResponsesPath: join(
           import.meta.dirname,
           'hooks-system.session-startup.responses',
         ),
         settings: {
-          hooks: {
+          hooksConfig: {
             enabled: true,
+          },
+          hooks: {
             SessionStart: [
               {
                 matcher: 'startup',
@@ -936,7 +1153,7 @@ console.log(JSON.stringify({
   }
 }));`;
 
-      await rig.setup('should fire SessionStart hook and inject context', {
+      rig.setup('should fire SessionStart hook and inject context', {
         fakeResponsesPath: join(
           import.meta.dirname,
           'hooks-system.session-startup.responses',
@@ -946,10 +1163,12 @@ console.log(JSON.stringify({
       const scriptPath = join(rig.testDir!, 'session_start_context_hook.cjs');
       writeFileSync(scriptPath, hookScript);
 
-      await rig.setup('should fire SessionStart hook and inject context', {
+      rig.setup('should fire SessionStart hook and inject context', {
         settings: {
-          hooks: {
+          hooksConfig: {
             enabled: true,
+          },
+          hooks: {
             SessionStart: [
               {
                 matcher: 'startup',
@@ -1011,7 +1230,7 @@ console.log(JSON.stringify({
   }
 }));`;
 
-      await rig.setup(
+      rig.setup(
         'should fire SessionStart hook and display systemMessage in interactive mode',
         {
           fakeResponsesPath: join(
@@ -1027,12 +1246,14 @@ console.log(JSON.stringify({
       );
       writeFileSync(scriptPath, hookScript);
 
-      await rig.setup(
+      rig.setup(
         'should fire SessionStart hook and display systemMessage in interactive mode',
         {
           settings: {
-            hooks: {
+            hooksConfig: {
               enabled: true,
+            },
+            hooks: {
               SessionStart: [
                 {
                   matcher: 'startup',
@@ -1091,7 +1312,7 @@ console.log(JSON.stringify({
       const sessionStartCommand =
         "node -e \"console.log(JSON.stringify({decision: 'allow', systemMessage: 'Session starting after clear'}))\"";
 
-      await rig.setup(
+      rig.setup(
         'should fire SessionEnd and SessionStart hooks on /clear command',
         {
           fakeResponsesPath: join(
@@ -1099,8 +1320,10 @@ console.log(JSON.stringify({
             'hooks-system.session-clear.responses',
           ),
           settings: {
-            hooks: {
+            hooksConfig: {
               enabled: true,
+            },
+            hooks: {
               SessionEnd: [
                 {
                   matcher: '*',
@@ -1265,14 +1488,16 @@ console.log(JSON.stringify({
       const preCompressCommand =
         "node -e \"console.log(JSON.stringify({decision: 'allow', systemMessage: 'PreCompress hook executed for automatic compression'}))\"";
 
-      await rig.setup('should fire PreCompress hook on automatic compression', {
+      rig.setup('should fire PreCompress hook on automatic compression', {
         fakeResponsesPath: join(
           import.meta.dirname,
           'hooks-system.compress-auto.responses',
         ),
         settings: {
-          hooks: {
+          hooksConfig: {
             enabled: true,
+          },
+          hooks: {
             PreCompress: [
               {
                 matcher: 'auto',
@@ -1330,14 +1555,16 @@ console.log(JSON.stringify({
       const sessionEndCommand =
         "node -e \"console.log(JSON.stringify({decision: 'allow', systemMessage: 'SessionEnd hook executed on exit'}))\"";
 
-      await rig.setup('should fire SessionEnd hook on graceful exit', {
+      rig.setup('should fire SessionEnd hook on graceful exit', {
         fakeResponsesPath: join(
           import.meta.dirname,
           'hooks-system.session-startup.responses',
         ),
         settings: {
-          hooks: {
+          hooksConfig: {
             enabled: true,
+          },
+          hooks: {
             SessionEnd: [
               {
                 matcher: 'exit',
@@ -1412,7 +1639,7 @@ console.log(JSON.stringify({
 
   describe('Hook Disabling', () => {
     it('should not execute hooks disabled in settings file', async () => {
-      await rig.setup('should not execute hooks disabled in settings file', {
+      rig.setup('should not execute hooks disabled in settings file', {
         fakeResponsesPath: join(
           import.meta.dirname,
           'hooks-system.disabled-via-settings.responses',
@@ -1432,10 +1659,13 @@ console.log(JSON.stringify({decision: "block", systemMessage: "Disabled hook sho
       writeFileSync(enabledPath, enabledHookScript);
       writeFileSync(disabledPath, disabledHookScript);
 
-      await rig.setup('should not execute hooks disabled in settings file', {
+      rig.setup('should not execute hooks disabled in settings file', {
         settings: {
-          hooks: {
+          hooksConfig: {
             enabled: true,
+            disabled: [`node "${disabledPath}"`], // Disable the second hook
+          },
+          hooks: {
             BeforeTool: [
               {
                 hooks: [
@@ -1452,7 +1682,6 @@ console.log(JSON.stringify({decision: "block", systemMessage: "Disabled hook sho
                 ],
               },
             ],
-            disabled: [`node "${disabledPath}"`], // Disable the second hook
           },
         },
       });
@@ -1487,15 +1716,12 @@ console.log(JSON.stringify({decision: "block", systemMessage: "Disabled hook sho
     });
 
     it('should respect disabled hooks across multiple operations', async () => {
-      await rig.setup(
-        'should respect disabled hooks across multiple operations',
-        {
-          fakeResponsesPath: join(
-            import.meta.dirname,
-            'hooks-system.disabled-via-command.responses',
-          ),
-        },
-      );
+      rig.setup('should respect disabled hooks across multiple operations', {
+        fakeResponsesPath: join(
+          import.meta.dirname,
+          'hooks-system.disabled-via-command.responses',
+        ),
+      });
 
       // Create two hook scripts - one that will be disabled, one that won't
       const activeHookScript = `const fs = require('fs');
@@ -1510,33 +1736,32 @@ console.log(JSON.stringify({decision: "block", systemMessage: "Disabled hook sho
       writeFileSync(activePath, activeHookScript);
       writeFileSync(disabledPath, disabledHookScript);
 
-      await rig.setup(
-        'should respect disabled hooks across multiple operations',
-        {
-          settings: {
-            hooks: {
-              enabled: true,
-              BeforeTool: [
-                {
-                  hooks: [
-                    {
-                      type: 'command',
-                      command: `node "${activePath}"`,
-                      timeout: 5000,
-                    },
-                    {
-                      type: 'command',
-                      command: `node "${disabledPath}"`,
-                      timeout: 5000,
-                    },
-                  ],
-                },
-              ],
-              disabled: [`node "${disabledPath}"`], // Disable the second hook
-            },
+      rig.setup('should respect disabled hooks across multiple operations', {
+        settings: {
+          hooksConfig: {
+            enabled: true,
+            disabled: [`node "${disabledPath}"`], // Disable the second hook,
+          },
+          hooks: {
+            BeforeTool: [
+              {
+                hooks: [
+                  {
+                    type: 'command',
+                    command: `node "${activePath}"`,
+                    timeout: 5000,
+                  },
+                  {
+                    type: 'command',
+                    command: `node "${disabledPath}"`,
+                    timeout: 5000,
+                  },
+                ],
+              },
+            ],
           },
         },
-      );
+      });
 
       // First run - only active hook should execute
       const result1 = await rig.run({
@@ -1587,9 +1812,7 @@ console.log(JSON.stringify({decision: "block", systemMessage: "Disabled hook sho
   describe('BeforeTool Hooks - Input Override', () => {
     it('should override tool input parameters via BeforeTool hook', async () => {
       // 1. First setup to get the test directory and prepare the hook script
-      await rig.setup(
-        'should override tool input parameters via BeforeTool hook',
-      );
+      rig.setup('should override tool input parameters via BeforeTool hook');
 
       // Create a hook script that overrides the tool input
       const hookOutput = {
@@ -1616,32 +1839,31 @@ console.log(JSON.stringify({decision: "block", systemMessage: "Disabled hook sho
       const commandPath = scriptPath.replace(/\\/g, '/');
 
       // 2. Full setup with settings and fake responses
-      await rig.setup(
-        'should override tool input parameters via BeforeTool hook',
-        {
-          fakeResponsesPath: join(
-            import.meta.dirname,
-            'hooks-system.input-modification.responses',
-          ),
-          settings: {
-            hooks: {
-              enabled: true,
-              BeforeTool: [
-                {
-                  matcher: 'write_file',
-                  hooks: [
-                    {
-                      type: 'command',
-                      command: `node "${commandPath}"`,
-                      timeout: 5000,
-                    },
-                  ],
-                },
-              ],
-            },
+      rig.setup('should override tool input parameters via BeforeTool hook', {
+        fakeResponsesPath: join(
+          import.meta.dirname,
+          'hooks-system.input-modification.responses',
+        ),
+        settings: {
+          hooksConfig: {
+            enabled: true,
+          },
+          hooks: {
+            BeforeTool: [
+              {
+                matcher: 'write_file',
+                hooks: [
+                  {
+                    type: 'command',
+                    command: `node "${commandPath}"`,
+                    timeout: 5000,
+                  },
+                ],
+              },
+            ],
           },
         },
-      );
+      });
 
       // Run the agent. The fake response will attempt to call write_file with
       // file_path="original.txt" and content="original content"
@@ -1698,19 +1920,21 @@ console.log(JSON.stringify({decision: "block", systemMessage: "Disabled hook sho
         hookOutput,
       )}));`;
 
-      await rig.setup('should stop agent execution via BeforeTool hook');
+      rig.setup('should stop agent execution via BeforeTool hook');
       const scriptPath = join(rig.testDir!, 'before_tool_stop_hook.js');
       writeFileSync(scriptPath, hookScript);
       const commandPath = scriptPath.replace(/\\/g, '/');
 
-      await rig.setup('should stop agent execution via BeforeTool hook', {
+      rig.setup('should stop agent execution via BeforeTool hook', {
         fakeResponsesPath: join(
           import.meta.dirname,
           'hooks-system.before-tool-stop.responses',
         ),
         settings: {
-          hooks: {
+          hooksConfig: {
             enabled: true,
+          },
+          hooks: {
             BeforeTool: [
               {
                 matcher: 'write_file',

integration-tests/run_shell_command.test.ts

@@ -164,7 +164,7 @@ describe('run_shell_command', () => {
     const result = await rig.run({
       args: [`--allowed-tools=run_shell_command(${tool})`],
       stdin: prompt,
-      yolo: false,
+      approvalMode: 'default',
     });
 
     const foundToolCall = await rig.waitForToolCall('run_shell_command', 15000);
@@ -207,7 +207,7 @@ describe('run_shell_command', () => {
     const result = await rig.run({
       args: '--allowed-tools=run_shell_command',
       stdin: prompt,
-      yolo: false,
+      approvalMode: 'default',
     });
 
     const foundToolCall = await rig.waitForToolCall('run_shell_command', 15000);
@@ -231,8 +231,8 @@ describe('run_shell_command', () => {
     expect(toolCall.toolRequest.success).toBe(true);
   });
 
-  it('should succeed with --yolo mode', async () => {
-    await rig.setup('should succeed with --yolo mode', {
+  it('should succeed in yolo mode', async () => {
+    await rig.setup('should succeed in yolo mode', {
       settings: { tools: { core: ['run_shell_command'] } },
     });
 
@@ -242,7 +242,7 @@ describe('run_shell_command', () => {
 
     const result = await rig.run({
       args: prompt,
-      yolo: true,
+      approvalMode: 'yolo',
     });
 
     const foundToolCall = await rig.waitForToolCall('run_shell_command', 15000);
@@ -276,7 +276,7 @@ describe('run_shell_command', () => {
     const result = await rig.run({
       args: `--allowed-tools=ShellTool(${tool})`,
       stdin: prompt,
-      yolo: false,
+      approvalMode: 'default',
     });
 
     const foundToolCall = await rig.waitForToolCall('run_shell_command', 15000);
@@ -325,7 +325,7 @@ describe('run_shell_command', () => {
         '--allowed-tools=run_shell_command(ls)',
       ],
       stdin: prompt,
-      yolo: false,
+      approvalMode: 'default',
     });
 
     for (const expected in ['ls', tool]) {
@@ -377,7 +377,7 @@ describe('run_shell_command', () => {
     const result = await rig.run({
       args: `--allowed-tools=run_shell_command(${allowedCommand})`,
       stdin: prompt,
-      yolo: false,
+      approvalMode: 'default',
     });
 
     if (!result.toLowerCase().includes('fail')) {
@@ -438,7 +438,7 @@ describe('run_shell_command', () => {
     await rig.run({
       args: `--allowed-tools=ShellTool(${chained.allowPattern})`,
       stdin: `${shellInjection}\n`,
-      yolo: false,
+      approvalMode: 'default',
     });
 
     // CLI should refuse to execute the chained command without scheduling run_shell_command.
@@ -470,7 +470,7 @@ describe('run_shell_command', () => {
         '--allowed-tools=run_shell_command',
       ],
       stdin: prompt,
-      yolo: false,
+      approvalMode: 'default',
     });
 
     const foundToolCall = await rig.waitForToolCall('run_shell_command', 15000);

integration-tests/save_memory.test.ts

@@ -1,54 +0,0 @@
-/**
- * @license
- * Copyright 2025 Google LLC
- * SPDX-License-Identifier: Apache-2.0
- */
-
-import { describe, it, expect, beforeEach, afterEach } from 'vitest';
-import { TestRig, printDebugInfo, validateModelOutput } from './test-helper.js';
-
-describe('save_memory', () => {
-  let rig: TestRig;
-
-  beforeEach(() => {
-    rig = new TestRig();
-  });
-
-  afterEach(async () => await rig.cleanup());
-
-  it('should be able to save to memory', async () => {
-    await rig.setup('should be able to save to memory', {
-      settings: { tools: { core: ['save_memory'] } },
-    });
-
-    const prompt = `remember that my favorite color is  blue.
-
-  what is my favorite color? tell me that and surround it with $ symbol`;
-    const result = await rig.run({ args: prompt });
-
-    const foundToolCall = await rig.waitForToolCall('save_memory');
-
-    // Add debugging information
-    if (!foundToolCall || !result.toLowerCase().includes('blue')) {
-      const allTools = printDebugInfo(rig, result, {
-        'Found tool call': foundToolCall,
-        'Contains blue': result.toLowerCase().includes('blue'),
-      });
-
-      console.error(
-        'Memory tool calls:',
-        allTools
-          .filter((t) => t.toolRequest.name === 'save_memory')
-          .map((t) => t.toolRequest.args),
-      );
-    }
-
-    expect(
-      foundToolCall,
-      'Expected to find a save_memory tool call',
-    ).toBeTruthy();
-
-    // Validate model output - will throw if no output, warn if missing expected content
-    validateModelOutput(result, 'blue', 'Save memory test');
-  });
-});

integration-tests/simple-mcp-server.test.ts

@@ -164,7 +164,7 @@ rpc.send({
 });
 `;
 
-describe('simple-mcp-server', () => {
+describe.skip('simple-mcp-server', () => {
   let rig: TestRig;
 
   beforeEach(() => {

integration-tests/skill-creator-scripts.test.ts

@@ -0,0 +1,104 @@
+/**
+ * @license
+ * Copyright 2026 Google LLC
+ * SPDX-License-Identifier: Apache-2.0
+ */
+
+import { describe, it, expect, beforeEach, afterEach } from 'vitest';
+import { TestRig } from './test-helper.js';
+import * as fs from 'node:fs';
+import * as path from 'node:path';
+import { execSync } from 'node:child_process';
+
+describe('skill-creator scripts e2e', () => {
+  let rig: TestRig;
+  const initScript = path.resolve(
+    'packages/core/src/skills/builtin/skill-creator/scripts/init_skill.cjs',
+  );
+  const validateScript = path.resolve(
+    'packages/core/src/skills/builtin/skill-creator/scripts/validate_skill.cjs',
+  );
+  const packageScript = path.resolve(
+    'packages/core/src/skills/builtin/skill-creator/scripts/package_skill.cjs',
+  );
+
+  beforeEach(() => {
+    rig = new TestRig();
+  });
+
+  afterEach(async () => {
+    await rig.cleanup();
+  });
+
+  it('should initialize, validate, and package a skill', async () => {
+    await rig.setup('skill-creator scripts e2e');
+    const skillName = 'e2e-test-skill';
+    const tempDir = rig.testDir!;
+
+    // 1. Initialize
+    execSync(`node "${initScript}" ${skillName} --path "${tempDir}"`, {
+      stdio: 'inherit',
+    });
+    const skillDir = path.join(tempDir, skillName);
+
+    expect(fs.existsSync(skillDir)).toBe(true);
+    expect(fs.existsSync(path.join(skillDir, 'SKILL.md'))).toBe(true);
+    expect(
+      fs.existsSync(path.join(skillDir, 'scripts/example_script.cjs')),
+    ).toBe(true);
+
+    // 2. Validate (should have warning initially due to TODOs)
+    const validateOutputInitial = execSync(
+      `node "${validateScript}" "${skillDir}" 2>&1`,
+      { encoding: 'utf8' },
+    );
+    expect(validateOutputInitial).toContain('⚠️  Found unresolved TODO');
+
+    // 3. Package (should fail due to TODOs)
+    try {
+      execSync(`node "${packageScript}" "${skillDir}" "${tempDir}"`, {
+        stdio: 'pipe',
+      });
+      throw new Error('Packaging should have failed due to TODOs');
+    } catch (err: unknown) {
+      expect((err as Error).message).toContain('Command failed');
+    }
+
+    // 4. Fix SKILL.md (remove TODOs)
+    let content = fs.readFileSync(path.join(skillDir, 'SKILL.md'), 'utf8');
+    // More aggressive global replace for all TODO patterns
+    content = content.replace(/TODO:[^\n]*/g, 'Fixed');
+    content = content.replace(/\[TODO:[^\]]*\]/g, 'Fixed');
+    fs.writeFileSync(path.join(skillDir, 'SKILL.md'), content);
+
+    // Also remove TODOs from example scripts
+    const exampleScriptPath = path.join(skillDir, 'scripts/example_script.cjs');
+    let scriptContent = fs.readFileSync(exampleScriptPath, 'utf8');
+    scriptContent = scriptContent.replace(/TODO:[^\n]*/g, 'Fixed');
+    fs.writeFileSync(exampleScriptPath, scriptContent);
+
+    // 4. Validate again (should pass now)
+    const validateOutput = execSync(`node "${validateScript}" "${skillDir}"`, {
+      encoding: 'utf8',
+    });
+    expect(validateOutput).toContain('Skill is valid!');
+
+    // 5. Package
+    execSync(`node "${packageScript}" "${skillDir}" "${tempDir}"`, {
+      stdio: 'inherit',
+    });
+    const skillFile = path.join(tempDir, `${skillName}.skill`);
+    expect(fs.existsSync(skillFile)).toBe(true);
+
+    // 6. Verify zip content (should NOT have nested directory)
+    // Use unzip -l if available, otherwise fallback to tar -tf (common on Windows)
+    let zipList: string;
+    try {
+      zipList = execSync(`unzip -l "${skillFile}"`, { encoding: 'utf8' });
+    } catch {
+      zipList = execSync(`tar -tf "${skillFile}"`, { encoding: 'utf8' });
+    }
+    expect(zipList).toContain('SKILL.md');
+    expect(zipList).not.toContain(`${skillName}/SKILL.md`);
+  });
+});

integration-tests/skill-creator-vulnerabilities.test.ts

@@ -0,0 +1,111 @@
+/**
+ * @license
+ * Copyright 2026 Google LLC
+ * SPDX-License-Identifier: Apache-2.0
+ */
+
+import { describe, it, expect, beforeEach, afterEach } from 'vitest';
+import { TestRig } from './test-helper.js';
+import * as fs from 'node:fs';
+import * as path from 'node:path';
+import { execSync, spawnSync } from 'node:child_process';
+
+describe('skill-creator scripts security and bug fixes', () => {
+  let rig: TestRig;
+  const initScript = path.resolve(
+    'packages/core/src/skills/builtin/skill-creator/scripts/init_skill.cjs',
+  );
+  const validateScript = path.resolve(
+    'packages/core/src/skills/builtin/skill-creator/scripts/validate_skill.cjs',
+  );
+  const packageScript = path.resolve(
+    'packages/core/src/skills/builtin/skill-creator/scripts/package_skill.cjs',
+  );
+
+  beforeEach(() => {
+    rig = new TestRig();
+  });
+
+  afterEach(async () => {
+    await rig.cleanup();
+  });
+
+  it('should prevent command injection in package_skill.cjs', async () => {
+    await rig.setup('skill-creator command injection');
+    const tempDir = rig.testDir!;
+
+    // Create a dummy skill
+    const skillName = 'injection-test';
+    execSync(`node "${initScript}" ${skillName} --path "${tempDir}"`);
+    const skillDir = path.join(tempDir, skillName);
+
+    // Malicious output filename with command injection
+    const maliciousFilename = '"; touch injection_success; #';
+
+    // Attempt to package with malicious filename
+    // We expect this to fail or at least NOT create the 'injection_success' file
+    spawnSync('node', [packageScript, skillDir, tempDir, maliciousFilename], {
+      cwd: tempDir,
+    });
+
+    const injectionFile = path.join(tempDir, 'injection_success');
+    expect(fs.existsSync(injectionFile)).toBe(false);
+  });
+
+  it('should prevent path traversal in init_skill.cjs', async () => {
+    await rig.setup('skill-creator init path traversal');
+    const tempDir = rig.testDir!;
+
+    const maliciousName = '../traversal-success';
+
+    const result = spawnSync(
+      'node',
+      [initScript, maliciousName, '--path', tempDir],
+      {
+        encoding: 'utf8',
+      },
+    );
+
+    expect(result.stderr).toContain(
+      'Error: Skill name cannot contain path separators',
+    );
+    const traversalDir = path.join(path.dirname(tempDir), 'traversal-success');
+    expect(fs.existsSync(traversalDir)).toBe(false);
+  });
+
+  it('should prevent path traversal in validate_skill.cjs', async () => {
+    await rig.setup('skill-creator validate path traversal');
+
+    const maliciousPath = '../../../../etc/passwd';
+    const result = spawnSync('node', [validateScript, maliciousPath], {
+      encoding: 'utf8',
+    });
+
+    expect(result.stderr).toContain('Error: Path traversal detected');
+  });
+
+  it('should not crash on empty description in validate_skill.cjs', async () => {
+    await rig.setup('skill-creator regex crash');
+    const tempDir = rig.testDir!;
+    const skillName = 'empty-desc-skill';
+
+    execSync(`node "${initScript}" ${skillName} --path "${tempDir}"`);
+    const skillDir = path.join(tempDir, skillName);
+    const skillMd = path.join(skillDir, 'SKILL.md');
+
+    // Set an empty quoted description
+    let content = fs.readFileSync(skillMd, 'utf8');
+    content = content.replace(/^description: .+$/m, 'description: ""');
+    fs.writeFileSync(skillMd, content);
+
+    const result = spawnSync('node', [validateScript, skillDir], {
+      encoding: 'utf8',
+    });
+
+    // It might still fail validation (e.g. TODOs), but it should NOT crash with a stack trace
+    expect(result.status).not.toBe(null);
+    expect(result.stderr).not.toContain(
+      "TypeError: Cannot read properties of undefined (reading 'trim')",
+    );
+  });
+});

package.json

@@ -1,6 +1,6 @@
 {
   "name": "@google/gemini-cli",
-  "version": "0.25.0-nightly.20260107.59a18e710",
+  "version": "0.27.0-nightly.20260121.97aac696f",
   "engines": {
     "node": ">=20.0.0"
   },
@@ -14,7 +14,7 @@
     "url": "git+https://github.com/google-gemini/gemini-cli.git"
   },
   "config": {
-    "sandboxImageUri": "us-docker.pkg.dev/gemini-code-dev/gemini-cli/sandbox:0.25.0-nightly.20260107.59a18e710"
+    "sandboxImageUri": "us-docker.pkg.dev/gemini-code-dev/gemini-cli/sandbox:0.27.0-nightly.20260121.97aac696f"
   },
   "scripts": {
     "start": "cross-env NODE_ENV=development node scripts/start.js",
@@ -41,6 +41,8 @@
     "test": "npm run test --workspaces --if-present",
     "test:ci": "npm run test:ci --workspaces --if-present && npm run test:scripts",
     "test:scripts": "vitest run --config ./scripts/tests/vitest.config.ts",
+    "test:always_passing_evals": "vitest run --config evals/vitest.config.ts",
+    "test:all_evals": "cross-env RUN_EVALS=1 vitest run --config evals/vitest.config.ts",
     "test:e2e": "cross-env VERBOSE=true KEEP_OUTPUT=true npm run test:integration:sandbox:none",
     "test:integration:all": "npm run test:integration:sandbox:none && npm run test:integration:sandbox:docker && npm run test:integration:sandbox:podman",
     "test:integration:sandbox:none": "cross-env GEMINI_SANDBOX=false vitest run --root ./integration-tests",
@@ -62,7 +64,7 @@
     "pre-commit": "node scripts/pre-commit.js"
   },
   "overrides": {
-    "ink": "npm:@jrichman/ink@6.4.6",
+    "ink": "npm:@jrichman/ink@6.4.8",
     "wrap-ansi": "9.0.2",
     "cliui": {
       "wrap-ansi": "7.0.0"
@@ -77,6 +79,7 @@
     "LICENSE"
   ],
   "devDependencies": {
+    "@agentclientprotocol/sdk": "^0.12.0",
     "@octokit/rest": "^22.0.0",
     "@types/marked": "^5.0.2",
     "@types/mime-types": "^3.0.1",
@@ -121,7 +124,7 @@
     "yargs": "^17.7.2"
   },
   "dependencies": {
-    "ink": "npm:@jrichman/ink@6.4.6",
+    "ink": "npm:@jrichman/ink@6.4.8",
     "latest-version": "^9.0.0",
     "simple-git": "^3.28.0"
   },

packages/a2a-server/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@google/gemini-cli-a2a-server",
-  "version": "0.25.0-nightly.20260107.59a18e710",
+  "version": "0.27.0-nightly.20260121.97aac696f",
   "description": "Gemini CLI A2A Server",
   "repository": {
     "type": "git",
@@ -25,7 +25,7 @@
     "dist"
   ],
   "dependencies": {
-    "@a2a-js/sdk": "^0.3.7",
+    "@a2a-js/sdk": "^0.3.8",
     "@google-cloud/storage": "^7.16.0",
     "@google/gemini-cli-core": "file:../core",
     "express": "^5.1.0",

packages/a2a-server/src/agent/task.test.ts

@@ -349,6 +349,44 @@ describe('Task', () => {
         }),
       );
     });
+
+    it.each([
+      { eventType: GeminiEventType.Retry, eventName: 'Retry' },
+      { eventType: GeminiEventType.InvalidStream, eventName: 'InvalidStream' },
+    ])(
+      'should handle $eventName event without triggering error handling',
+      async ({ eventType }) => {
+        const mockConfig = createMockConfig();
+        const mockEventBus: ExecutionEventBus = {
+          publish: vi.fn(),
+          on: vi.fn(),
+          off: vi.fn(),
+          once: vi.fn(),
+          removeAllListeners: vi.fn(),
+          finished: vi.fn(),
+        };
+
+        // @ts-expect-error - Calling private constructor
+        const task = new Task(
+          'task-id',
+          'context-id',
+          mockConfig as Config,
+          mockEventBus,
+        );
+
+        const cancelPendingToolsSpy = vi.spyOn(task, 'cancelPendingTools');
+        const setTaskStateSpy = vi.spyOn(task, 'setTaskStateAndPublishUpdate');
+
+        const event = {
+          type: eventType,
+        };
+
+        await task.acceptAgentMessage(event);
+
+        expect(cancelPendingToolsSpy).not.toHaveBeenCalled();
+        expect(setTaskStateSpy).not.toHaveBeenCalled();
+      },
+    );
   });
 
   describe('_schedulerToolCallsUpdate', () => {

packages/a2a-server/src/agent/task.ts

@@ -378,7 +378,7 @@ export class Task {
       if (tc.status === 'awaiting_approval' && tc.confirmationDetails) {
         this.pendingToolConfirmationDetails.set(
           tc.request.callId,
-          tc.confirmationDetails,
+          tc.confirmationDetails as ToolCallConfirmationDetails,
         );
       }
 
@@ -412,7 +412,9 @@ export class Task {
       toolCalls.forEach((tc: ToolCall) => {
         if (tc.status === 'awaiting_approval' && tc.confirmationDetails) {
           // eslint-disable-next-line @typescript-eslint/no-floating-promises
-          tc.confirmationDetails.onConfirm(ToolConfirmationOutcome.ProceedOnce);
+          (tc.confirmationDetails as ToolCallConfirmationDetails).onConfirm(
+            ToolConfirmationOutcome.ProceedOnce,
+          );
           this.pendingToolConfirmationDetails.delete(tc.request.callId);
         }
       });
@@ -573,7 +575,10 @@ export class Task {
       EDIT_TOOL_NAMES.has(request.name),
     );
 
-    if (restorableToolCalls.length > 0) {
+    if (
+      restorableToolCalls.length > 0 &&
+      this.config.getCheckpointingEnabled()
+    ) {
       const gitService = await this.config.getGitService();
       if (gitService) {
         const { checkpointsToWrite, toolCallToCheckpointMap, errors } =
@@ -707,6 +712,10 @@ export class Task {
       case GeminiEventType.ModelInfo:
         this.modelInfo = event.value;
         break;
+      case GeminiEventType.Retry:
+      case GeminiEventType.InvalidStream:
+        // An invalid stream should trigger a retry, which requires no action from the user.
+        break;
       case GeminiEventType.Error:
       default: {
         // Block scope for lexical declaration

packages/a2a-server/src/commands/command-registry.ts

@@ -4,6 +4,7 @@
  * SPDX-License-Identifier: Apache-2.0
  */
 
+import { MemoryCommand } from './memory.js';
 import { debugLogger } from '@google/gemini-cli-core';
 import { ExtensionsCommand } from './extensions.js';
 import { InitCommand } from './init.js';
@@ -22,6 +23,7 @@ export class CommandRegistry {
     this.register(new ExtensionsCommand());
     this.register(new RestoreCommand());
     this.register(new InitCommand());
+    this.register(new MemoryCommand());
   }
 
   register(command: Command) {

packages/a2a-server/src/commands/init.test.ts

@@ -26,10 +26,14 @@ vi.mock('@google/gemini-cli-core', async (importOriginal) => {
   };
 });
 
-vi.mock('node:fs', () => ({
-  existsSync: vi.fn(),
-  writeFileSync: vi.fn(),
-}));
+vi.mock('node:fs', async (importOriginal) => {
+  const actual = await importOriginal<typeof import('node:fs')>();
+  return {
+    ...actual,
+    existsSync: vi.fn(),
+    writeFileSync: vi.fn(),
+  };
+});
 
 vi.mock('../agent/executor.js', () => ({
   CoderAgentExecutor: vi.fn().mockImplementation(() => ({

packages/a2a-server/src/commands/memory.test.ts

@@ -0,0 +1,208 @@
+/**
+ * @license
+ * Copyright 2025 Google LLC
+ * SPDX-License-Identifier: Apache-2.0
+ */
+
+import {
+  addMemory,
+  listMemoryFiles,
+  refreshMemory,
+  showMemory,
+} from '@google/gemini-cli-core';
+import { beforeEach, describe, expect, it, vi } from 'vitest';
+import {
+  AddMemoryCommand,
+  ListMemoryCommand,
+  MemoryCommand,
+  RefreshMemoryCommand,
+  ShowMemoryCommand,
+} from './memory.js';
+import type { CommandContext } from './types.js';
+import type {
+  AnyDeclarativeTool,
+  Config,
+  ToolRegistry,
+} from '@google/gemini-cli-core';
+
+// Mock the core functions
+vi.mock('@google/gemini-cli-core', async (importOriginal) => {
+  const actual =
+    await importOriginal<typeof import('@google/gemini-cli-core')>();
+  return {
+    ...actual,
+    showMemory: vi.fn(),
+    refreshMemory: vi.fn(),
+    listMemoryFiles: vi.fn(),
+    addMemory: vi.fn(),
+  };
+});
+
+const mockShowMemory = vi.mocked(showMemory);
+const mockRefreshMemory = vi.mocked(refreshMemory);
+const mockListMemoryFiles = vi.mocked(listMemoryFiles);
+const mockAddMemory = vi.mocked(addMemory);
+
+describe('a2a-server memory commands', () => {
+  let mockContext: CommandContext;
+  let mockConfig: Config;
+  let mockToolRegistry: ToolRegistry;
+  let mockSaveMemoryTool: AnyDeclarativeTool;
+
+  beforeEach(() => {
+    mockSaveMemoryTool = {
+      name: 'save_memory',
+      description: 'Saves memory',
+      buildAndExecute: vi.fn().mockResolvedValue(undefined),
+    } as unknown as AnyDeclarativeTool;
+
+    mockToolRegistry = {
+      getTool: vi.fn(),
+    } as unknown as ToolRegistry;
+
+    mockConfig = {
+      getToolRegistry: vi.fn().mockReturnValue(mockToolRegistry),
+    } as unknown as Config;
+
+    mockContext = {
+      config: mockConfig,
+    };
+
+    vi.mocked(mockToolRegistry.getTool).mockReturnValue(mockSaveMemoryTool);
+  });
+
+  describe('MemoryCommand', () => {
+    it('delegates to ShowMemoryCommand', async () => {
+      const command = new MemoryCommand();
+      mockShowMemory.mockReturnValue({
+        type: 'message',
+        messageType: 'info',
+        content: 'showing memory',
+      });
+      const response = await command.execute(mockContext, []);
+      expect(response.data).toBe('showing memory');
+      expect(mockShowMemory).toHaveBeenCalledWith(mockContext.config);
+    });
+  });
+
+  describe('ShowMemoryCommand', () => {
+    it('executes showMemory and returns the content', async () => {
+      const command = new ShowMemoryCommand();
+      mockShowMemory.mockReturnValue({
+        type: 'message',
+        messageType: 'info',
+        content: 'test memory content',
+      });
+
+      const response = await command.execute(mockContext, []);
+
+      expect(mockShowMemory).toHaveBeenCalledWith(mockContext.config);
+      expect(response.name).toBe('memory show');
+      expect(response.data).toBe('test memory content');
+    });
+  });
+
+  describe('RefreshMemoryCommand', () => {
+    it('executes refreshMemory and returns the content', async () => {
+      const command = new RefreshMemoryCommand();
+      mockRefreshMemory.mockResolvedValue({
+        type: 'message',
+        messageType: 'info',
+        content: 'memory refreshed',
+      });
+
+      const response = await command.execute(mockContext, []);
+
+      expect(mockRefreshMemory).toHaveBeenCalledWith(mockContext.config);
+      expect(response.name).toBe('memory refresh');
+      expect(response.data).toBe('memory refreshed');
+    });
+  });
+
+  describe('ListMemoryCommand', () => {
+    it('executes listMemoryFiles and returns the content', async () => {
+      const command = new ListMemoryCommand();
+      mockListMemoryFiles.mockReturnValue({
+        type: 'message',
+        messageType: 'info',
+        content: 'file1.md\nfile2.md',
+      });
+
+      const response = await command.execute(mockContext, []);
+
+      expect(mockListMemoryFiles).toHaveBeenCalledWith(mockContext.config);
+      expect(response.name).toBe('memory list');
+      expect(response.data).toBe('file1.md\nfile2.md');
+    });
+  });
+
+  describe('AddMemoryCommand', () => {
+    it('returns message content if addMemory returns a message', async () => {
+      const command = new AddMemoryCommand();
+      mockAddMemory.mockReturnValue({
+        type: 'message',
+        messageType: 'error',
+        content: 'error message',
+      });
+
+      const response = await command.execute(mockContext, []);
+
+      expect(mockAddMemory).toHaveBeenCalledWith('');
+      expect(response.name).toBe('memory add');
+      expect(response.data).toBe('error message');
+    });
+
+    it('executes the save_memory tool if found', async () => {
+      const command = new AddMemoryCommand();
+      const fact = 'this is a new fact';
+      mockAddMemory.mockReturnValue({
+        type: 'tool',
+        toolName: 'save_memory',
+        toolArgs: { fact },
+      });
+
+      const response = await command.execute(mockContext, [
+        'this',
+        'is',
+        'a',
+        'new',
+        'fact',
+      ]);
+
+      expect(mockAddMemory).toHaveBeenCalledWith(fact);
+      expect(mockConfig.getToolRegistry).toHaveBeenCalled();
+      expect(mockToolRegistry.getTool).toHaveBeenCalledWith('save_memory');
+      expect(mockSaveMemoryTool.buildAndExecute).toHaveBeenCalledWith(
+        { fact },
+        expect.any(AbortSignal),
+        undefined,
+        {
+          sanitizationConfig: {
+            allowedEnvironmentVariables: [],
+            blockedEnvironmentVariables: [],
+            enableEnvironmentVariableRedaction: false,
+          },
+        },
+      );
+      expect(mockRefreshMemory).toHaveBeenCalledWith(mockContext.config);
+      expect(response.name).toBe('memory add');
+      expect(response.data).toBe(`Added memory: "${fact}"`);
+    });
+
+    it('returns an error if the tool is not found', async () => {
+      const command = new AddMemoryCommand();
+      const fact = 'another fact';
+      mockAddMemory.mockReturnValue({
+        type: 'tool',
+        toolName: 'save_memory',
+        toolArgs: { fact },
+      });
+      vi.mocked(mockToolRegistry.getTool).mockReturnValue(undefined);
+
+      const response = await command.execute(mockContext, ['another', 'fact']);
+
+      expect(response.name).toBe('memory add');
+      expect(response.data).toBe('Error: Tool save_memory not found.');
+    });
+  });
+});