feat(core): add concurrency safety guidance for subagent delegation (#17753) (#21278)

evals/concurrency-safety.eval.ts

@@ -0,0 +1,56 @@
+/**
+ * @license
+ * Copyright 2026 Google LLC
+ * SPDX-License-Identifier: Apache-2.0
+ */
+
+import { expect } from 'vitest';
+import { evalTest } from './test-helper.js';
+
+const MUTATION_AGENT_DEFINITION = `---
+name: mutation-agent
+description: An agent that modifies the workspace (writes, deletes, git operations, etc).
+max_turns: 1
+tools:
+  - write_file
+---
+
+You are the mutation agent. Do the mutation requested.
+`;
+
+describe('concurrency safety eval test cases', () => {
+  evalTest('USUALLY_PASSES', {
+    name: 'mutation agents are run in parallel when explicitly requested',
+    params: {
+      settings: {
+        experimental: {
+          enableAgents: true,
+        },
+      },
+    },
+    prompt:
+      'Update A.txt to say "A" and update B.txt to say "B". Delegate these tasks to two separate mutation-agent subagents. You MUST run these subagents in parallel at the same time.',
+    files: {
+      '.gemini/agents/mutation-agent.md': MUTATION_AGENT_DEFINITION,
+    },
+    assert: async (rig) => {
+      const logs = rig.readToolLogs();
+      const mutationCalls = logs.filter(
+        (log) => log.toolRequest?.name === 'mutation-agent',
+      );
+
+      expect(
+        mutationCalls.length,
+        'Agent should have called the mutation-agent at least twice',
+      ).toBeGreaterThanOrEqual(2);
+
+      const firstPromptId = mutationCalls[0].toolRequest.prompt_id;
+      const secondPromptId = mutationCalls[1].toolRequest.prompt_id;
+
+      expect(
+        firstPromptId,
+        'mutation agents should be called in parallel (same turn / prompt_ids) when explicitly requested',
+      ).toEqual(secondPromptId);
+    },
+  });
+});

evals/test-helper.ts

@@ -112,6 +112,7 @@ export function evalTest(policy: EvalPolicy, evalCase: EvalCase) {
         // commands.
         execSync('git config core.editor "true"', execOptions);
         execSync('git config core.pager "cat"', execOptions);
+        execSync('git config commit.gpgsign false', execOptions);
         execSync('git add .', execOptions);
         execSync('git commit --allow-empty -m "Initial commit"', execOptions);
       }

packages/core/src/core/__snapshots__/prompts.test.ts.snap

@@ -62,6 +62,8 @@ Operate as a **strategic orchestrator**. Your own context window is your most pr
 
 When you delegate, the sub-agent's entire execution is consolidated into a single summary in your history, keeping your main loop lean.
 
+**Concurrency Safety and Mandate:** You should NEVER run multiple subagents in a single turn if their abilities mutate the same files or resources. This is to prevent race conditions and ensure that the workspace is in a consistent state. Only run multiple subagents in parallel when their tasks are independent (e.g., multiple concurrent research or read-only tasks) or if parallel execution is explicitly requested by the user.
+
 **High-Impact Delegation Candidates:**
 - **Repetitive Batch Tasks:** Tasks involving more than 3 files or repeated steps (e.g., "Add license headers to all files in src/", "Fix all lint errors in the project").
 - **High-Volume Output:** Commands or tools expected to return large amounts of data (e.g., verbose builds, exhaustive file searches).
@@ -230,6 +232,8 @@ Operate as a **strategic orchestrator**. Your own context window is your most pr
 
 When you delegate, the sub-agent's entire execution is consolidated into a single summary in your history, keeping your main loop lean.
 
+**Concurrency Safety and Mandate:** You should NEVER run multiple subagents in a single turn if their abilities mutate the same files or resources. This is to prevent race conditions and ensure that the workspace is in a consistent state. Only run multiple subagents in parallel when their tasks are independent (e.g., multiple concurrent research or read-only tasks) or if parallel execution is explicitly requested by the user.
+
 **High-Impact Delegation Candidates:**
 - **Repetitive Batch Tasks:** Tasks involving more than 3 files or repeated steps (e.g., "Add license headers to all files in src/", "Fix all lint errors in the project").
 - **High-Volume Output:** Commands or tools expected to return large amounts of data (e.g., verbose builds, exhaustive file searches).
@@ -517,6 +521,8 @@ Operate as a **strategic orchestrator**. Your own context window is your most pr
 
 When you delegate, the sub-agent's entire execution is consolidated into a single summary in your history, keeping your main loop lean.
 
+**Concurrency Safety and Mandate:** You should NEVER run multiple subagents in a single turn if their abilities mutate the same files or resources. This is to prevent race conditions and ensure that the workspace is in a consistent state. Only run multiple subagents in parallel when their tasks are independent (e.g., multiple concurrent research or read-only tasks) or if parallel execution is explicitly requested by the user.
+
 **High-Impact Delegation Candidates:**
 - **Repetitive Batch Tasks:** Tasks involving more than 3 files or repeated steps (e.g., "Add license headers to all files in src/", "Fix all lint errors in the project").
 - **High-Volume Output:** Commands or tools expected to return large amounts of data (e.g., verbose builds, exhaustive file searches).
@@ -685,6 +691,8 @@ Operate as a **strategic orchestrator**. Your own context window is your most pr
 
 When you delegate, the sub-agent's entire execution is consolidated into a single summary in your history, keeping your main loop lean.
 
+**Concurrency Safety and Mandate:** You should NEVER run multiple subagents in a single turn if their abilities mutate the same files or resources. This is to prevent race conditions and ensure that the workspace is in a consistent state. Only run multiple subagents in parallel when their tasks are independent (e.g., multiple concurrent research or read-only tasks) or if parallel execution is explicitly requested by the user.
+
 **High-Impact Delegation Candidates:**
 - **Repetitive Batch Tasks:** Tasks involving more than 3 files or repeated steps (e.g., "Add license headers to all files in src/", "Fix all lint errors in the project").
 - **High-Volume Output:** Commands or tools expected to return large amounts of data (e.g., verbose builds, exhaustive file searches).
@@ -1571,6 +1579,8 @@ Operate as a **strategic orchestrator**. Your own context window is your most pr
 
 When you delegate, the sub-agent's entire execution is consolidated into a single summary in your history, keeping your main loop lean.
 
+**Concurrency Safety and Mandate:** You should NEVER run multiple subagents in a single turn if their abilities mutate the same files or resources. This is to prevent race conditions and ensure that the workspace is in a consistent state. Only run multiple subagents in parallel when their tasks are independent (e.g., multiple concurrent research or read-only tasks) or if parallel execution is explicitly requested by the user.
+
 **High-Impact Delegation Candidates:**
 - **Repetitive Batch Tasks:** Tasks involving more than 3 files or repeated steps (e.g., "Add license headers to all files in src/", "Fix all lint errors in the project").
 - **High-Volume Output:** Commands or tools expected to return large amounts of data (e.g., verbose builds, exhaustive file searches).
@@ -1734,6 +1744,8 @@ Operate as a **strategic orchestrator**. Your own context window is your most pr
 
 When you delegate, the sub-agent's entire execution is consolidated into a single summary in your history, keeping your main loop lean.
 
+**Concurrency Safety and Mandate:** You should NEVER run multiple subagents in a single turn if their abilities mutate the same files or resources. This is to prevent race conditions and ensure that the workspace is in a consistent state. Only run multiple subagents in parallel when their tasks are independent (e.g., multiple concurrent research or read-only tasks) or if parallel execution is explicitly requested by the user.
+
 **High-Impact Delegation Candidates:**
 - **Repetitive Batch Tasks:** Tasks involving more than 3 files or repeated steps (e.g., "Add license headers to all files in src/", "Fix all lint errors in the project").
 - **High-Volume Output:** Commands or tools expected to return large amounts of data (e.g., verbose builds, exhaustive file searches).
@@ -1889,6 +1901,8 @@ Operate as a **strategic orchestrator**. Your own context window is your most pr
 
 When you delegate, the sub-agent's entire execution is consolidated into a single summary in your history, keeping your main loop lean.
 
+**Concurrency Safety and Mandate:** You should NEVER run multiple subagents in a single turn if their abilities mutate the same files or resources. This is to prevent race conditions and ensure that the workspace is in a consistent state. Only run multiple subagents in parallel when their tasks are independent (e.g., multiple concurrent research or read-only tasks) or if parallel execution is explicitly requested by the user.
+
 **High-Impact Delegation Candidates:**
 - **Repetitive Batch Tasks:** Tasks involving more than 3 files or repeated steps (e.g., "Add license headers to all files in src/", "Fix all lint errors in the project").
 - **High-Volume Output:** Commands or tools expected to return large amounts of data (e.g., verbose builds, exhaustive file searches).
@@ -2044,6 +2058,8 @@ Operate as a **strategic orchestrator**. Your own context window is your most pr
 
 When you delegate, the sub-agent's entire execution is consolidated into a single summary in your history, keeping your main loop lean.
 
+**Concurrency Safety and Mandate:** You should NEVER run multiple subagents in a single turn if their abilities mutate the same files or resources. This is to prevent race conditions and ensure that the workspace is in a consistent state. Only run multiple subagents in parallel when their tasks are independent (e.g., multiple concurrent research or read-only tasks) or if parallel execution is explicitly requested by the user.
+
 **High-Impact Delegation Candidates:**
 - **Repetitive Batch Tasks:** Tasks involving more than 3 files or repeated steps (e.g., "Add license headers to all files in src/", "Fix all lint errors in the project").
 - **High-Volume Output:** Commands or tools expected to return large amounts of data (e.g., verbose builds, exhaustive file searches).
@@ -2195,6 +2211,8 @@ Operate as a **strategic orchestrator**. Your own context window is your most pr
 
 When you delegate, the sub-agent's entire execution is consolidated into a single summary in your history, keeping your main loop lean.
 
+**Concurrency Safety and Mandate:** You should NEVER run multiple subagents in a single turn if their abilities mutate the same files or resources. This is to prevent race conditions and ensure that the workspace is in a consistent state. Only run multiple subagents in parallel when their tasks are independent (e.g., multiple concurrent research or read-only tasks) or if parallel execution is explicitly requested by the user.
+
 **High-Impact Delegation Candidates:**
 - **Repetitive Batch Tasks:** Tasks involving more than 3 files or repeated steps (e.g., "Add license headers to all files in src/", "Fix all lint errors in the project").
 - **High-Volume Output:** Commands or tools expected to return large amounts of data (e.g., verbose builds, exhaustive file searches).
@@ -2346,6 +2364,8 @@ Operate as a **strategic orchestrator**. Your own context window is your most pr
 
 When you delegate, the sub-agent's entire execution is consolidated into a single summary in your history, keeping your main loop lean.
 
+**Concurrency Safety and Mandate:** You should NEVER run multiple subagents in a single turn if their abilities mutate the same files or resources. This is to prevent race conditions and ensure that the workspace is in a consistent state. Only run multiple subagents in parallel when their tasks are independent (e.g., multiple concurrent research or read-only tasks) or if parallel execution is explicitly requested by the user.
+
 **High-Impact Delegation Candidates:**
 - **Repetitive Batch Tasks:** Tasks involving more than 3 files or repeated steps (e.g., "Add license headers to all files in src/", "Fix all lint errors in the project").
 - **High-Volume Output:** Commands or tools expected to return large amounts of data (e.g., verbose builds, exhaustive file searches).
@@ -2489,6 +2509,8 @@ Operate as a **strategic orchestrator**. Your own context window is your most pr
 
 When you delegate, the sub-agent's entire execution is consolidated into a single summary in your history, keeping your main loop lean.
 
+**Concurrency Safety and Mandate:** You should NEVER run multiple subagents in a single turn if their abilities mutate the same files or resources. This is to prevent race conditions and ensure that the workspace is in a consistent state. Only run multiple subagents in parallel when their tasks are independent (e.g., multiple concurrent research or read-only tasks) or if parallel execution is explicitly requested by the user.
+
 **High-Impact Delegation Candidates:**
 - **Repetitive Batch Tasks:** Tasks involving more than 3 files or repeated steps (e.g., "Add license headers to all files in src/", "Fix all lint errors in the project").
 - **High-Volume Output:** Commands or tools expected to return large amounts of data (e.g., verbose builds, exhaustive file searches).
@@ -2639,6 +2661,8 @@ Operate as a **strategic orchestrator**. Your own context window is your most pr
 
 When you delegate, the sub-agent's entire execution is consolidated into a single summary in your history, keeping your main loop lean.
 
+**Concurrency Safety and Mandate:** You should NEVER run multiple subagents in a single turn if their abilities mutate the same files or resources. This is to prevent race conditions and ensure that the workspace is in a consistent state. Only run multiple subagents in parallel when their tasks are independent (e.g., multiple concurrent research or read-only tasks) or if parallel execution is explicitly requested by the user.
+
 **High-Impact Delegation Candidates:**
 - **Repetitive Batch Tasks:** Tasks involving more than 3 files or repeated steps (e.g., "Add license headers to all files in src/", "Fix all lint errors in the project").
 - **High-Volume Output:** Commands or tools expected to return large amounts of data (e.g., verbose builds, exhaustive file searches).
@@ -2790,6 +2814,8 @@ Operate as a **strategic orchestrator**. Your own context window is your most pr
 
 When you delegate, the sub-agent's entire execution is consolidated into a single summary in your history, keeping your main loop lean.
 
+**Concurrency Safety and Mandate:** You should NEVER run multiple subagents in a single turn if their abilities mutate the same files or resources. This is to prevent race conditions and ensure that the workspace is in a consistent state. Only run multiple subagents in parallel when their tasks are independent (e.g., multiple concurrent research or read-only tasks) or if parallel execution is explicitly requested by the user.
+
 **High-Impact Delegation Candidates:**
 - **Repetitive Batch Tasks:** Tasks involving more than 3 files or repeated steps (e.g., "Add license headers to all files in src/", "Fix all lint errors in the project").
 - **High-Volume Output:** Commands or tools expected to return large amounts of data (e.g., verbose builds, exhaustive file searches).
@@ -3193,6 +3219,8 @@ Operate as a **strategic orchestrator**. Your own context window is your most pr
 
 When you delegate, the sub-agent's entire execution is consolidated into a single summary in your history, keeping your main loop lean.
 
+**Concurrency Safety and Mandate:** You should NEVER run multiple subagents in a single turn if their abilities mutate the same files or resources. This is to prevent race conditions and ensure that the workspace is in a consistent state. Only run multiple subagents in parallel when their tasks are independent (e.g., multiple concurrent research or read-only tasks) or if parallel execution is explicitly requested by the user.
+
 **High-Impact Delegation Candidates:**
 - **Repetitive Batch Tasks:** Tasks involving more than 3 files or repeated steps (e.g., "Add license headers to all files in src/", "Fix all lint errors in the project").
 - **High-Volume Output:** Commands or tools expected to return large amounts of data (e.g., verbose builds, exhaustive file searches).
@@ -3344,6 +3372,8 @@ Operate as a **strategic orchestrator**. Your own context window is your most pr
 
 When you delegate, the sub-agent's entire execution is consolidated into a single summary in your history, keeping your main loop lean.
 
+**Concurrency Safety and Mandate:** You should NEVER run multiple subagents in a single turn if their abilities mutate the same files or resources. This is to prevent race conditions and ensure that the workspace is in a consistent state. Only run multiple subagents in parallel when their tasks are independent (e.g., multiple concurrent research or read-only tasks) or if parallel execution is explicitly requested by the user.
+
 **High-Impact Delegation Candidates:**
 - **Repetitive Batch Tasks:** Tasks involving more than 3 files or repeated steps (e.g., "Add license headers to all files in src/", "Fix all lint errors in the project").
 - **High-Volume Output:** Commands or tools expected to return large amounts of data (e.g., verbose builds, exhaustive file searches).
@@ -3607,6 +3637,8 @@ Operate as a **strategic orchestrator**. Your own context window is your most pr
 
 When you delegate, the sub-agent's entire execution is consolidated into a single summary in your history, keeping your main loop lean.
 
+**Concurrency Safety and Mandate:** You should NEVER run multiple subagents in a single turn if their abilities mutate the same files or resources. This is to prevent race conditions and ensure that the workspace is in a consistent state. Only run multiple subagents in parallel when their tasks are independent (e.g., multiple concurrent research or read-only tasks) or if parallel execution is explicitly requested by the user.
+
 **High-Impact Delegation Candidates:**
 - **Repetitive Batch Tasks:** Tasks involving more than 3 files or repeated steps (e.g., "Add license headers to all files in src/", "Fix all lint errors in the project").
 - **High-Volume Output:** Commands or tools expected to return large amounts of data (e.g., verbose builds, exhaustive file searches).
@@ -3758,6 +3790,8 @@ Operate as a **strategic orchestrator**. Your own context window is your most pr
 
 When you delegate, the sub-agent's entire execution is consolidated into a single summary in your history, keeping your main loop lean.
 
+**Concurrency Safety and Mandate:** You should NEVER run multiple subagents in a single turn if their abilities mutate the same files or resources. This is to prevent race conditions and ensure that the workspace is in a consistent state. Only run multiple subagents in parallel when their tasks are independent (e.g., multiple concurrent research or read-only tasks) or if parallel execution is explicitly requested by the user.
+
 **High-Impact Delegation Candidates:**
 - **Repetitive Batch Tasks:** Tasks involving more than 3 files or repeated steps (e.g., "Add license headers to all files in src/", "Fix all lint errors in the project").
 - **High-Volume Output:** Commands or tools expected to return large amounts of data (e.g., verbose builds, exhaustive file searches).

packages/core/src/prompts/snippets.ts

@@ -251,6 +251,8 @@ Operate as a **strategic orchestrator**. Your own context window is your most pr
 
 When you delegate, the sub-agent's entire execution is consolidated into a single summary in your history, keeping your main loop lean.
 
+**Concurrency Safety and Mandate:** You should NEVER run multiple subagents in a single turn if their abilities mutate the same files or resources. This is to prevent race conditions and ensure that the workspace is in a consistent state. Only run multiple subagents in parallel when their tasks are independent (e.g., multiple concurrent research or read-only tasks) or if parallel execution is explicitly requested by the user.
+
 **High-Impact Delegation Candidates:**
 - **Repetitive Batch Tasks:** Tasks involving more than 3 files or repeated steps (e.g., "Add license headers to all files in src/", "Fix all lint errors in the project").
 - **High-Volume Output:** Commands or tools expected to return large amounts of data (e.g., verbose builds, exhaustive file searches).