8d584e4d96
fix(core): handle nested plan files by resolving paths correctly
...
This fixes a bug where path.basename was incorrectly stripping directory structures from plan files (e.g., trying to write to plans/nested/file.md would incorrectly write to plans/file.md). By using path.resolve and verifying with isSubpath, nested files are now handled securely and correctly.
2026-04-10 18:33:02 +00:00
a99cd0be28
Merge branch 'main' into worktree-con-plan-bug
...
Resolves conflict in packages/core/src/tools/enter-plan-mode.test.ts by removing an assertion for directory creation, which has been centralized in config.ts in this branch.
2026-04-10 18:00:53 +00:00
74ce3eef0c
fix(evals): refine plan mode eval prompt to ensure toolchain completion
...
Align the 'foo' test prompt with existing project conventions while ensuring the model has the 'informal agreement' signal required to proceed to formal approval and implementation.
2026-04-09 18:13:35 +00:00
b238a453e3
feat(core): refactor subagent tool to unified invoke_subagent tool ( #24489 )
2026-04-09 16:48:24 +00:00
f1bb2af6de
Generalize evals infra to support more types of evals, organization and queuing of named suites ( #24941 )
2026-04-08 23:57:26 +00:00
d2b775f9a7
Add an eval for and fix unsafe cloning behavior. ( #24457 )
2026-04-07 03:17:44 +00:00
8f131ffef7
Fix issue where topic headers can be posted back to back ( #24759 )
2026-04-06 18:36:22 +00:00
4fb3790051
feat(core): discourage update topic tool for simple tasks ( #24640 )
...
Co-authored-by: Samee Zahid <sameez@google.com >
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
2026-04-03 23:52:24 +00:00
f510394721
Implement background process monitoring and inspection tools ( #23799 )
2026-04-02 15:01:00 +00:00
973092df50
feat: implement high-signal PR regression check for evaluations ( #23937 )
2026-04-02 05:14:43 +00:00
43cf63e189
fix: update task tracker storage location in system prompt ( #24034 )
2026-04-01 18:29:09 +00:00
ca43f8c291
feat(core): prioritize discussion before formal plan approval ( #24423 )
2026-04-01 15:55:47 +00:00
94f9480a3a
fix(core): resolve Plan Mode deadlock during plan file creation due to sandbox restrictions ( #24047 )
2026-03-31 22:06:50 +00:00
07e2053e12
feat(plan): promote planning feature to stable ( #24282 )
2026-03-31 16:10:13 +00:00
117a2d3844
fix(evals): add update_topic behavioral eval ( #24223 )
2026-03-30 22:02:53 +00:00
d9d2ce36f2
test(evals): add comprehensive subagent delegation evaluations ( #24132 )
2026-03-29 23:13:50 +00:00
2e03e3aed5
feat(evals): add reliability harvester and 500/503 retry support ( #23626 )
2026-03-26 01:48:45 +00:00
84f40768a1
feat(evals): centralize test agents into test-utils for reuse ( #23616 )
...
Co-authored-by: Samee Zahid <sameez@google.com >
2026-03-24 19:50:48 +00:00
bf80e27dbc
test(evals): fix overlapping act() deadlock in app-test-helper ( #23666 )
2026-03-24 19:12:22 +00:00
6b7dc4d822
refactor(core): stop gemini CLI from producing unsafe casts ( #23611 )
2026-03-24 16:19:59 +00:00
dcedc42979
fix(plan): sandbox path resolution in Plan Mode to prevent hallucinations ( #22737 )
...
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
2026-03-24 13:19:29 +00:00
36e6445dba
feat(sandbox): dynamic macOS sandbox expansion and worktree support ( #23301 )
2026-03-24 04:48:13 +00:00
57a66f5f0d
feat(evals): add behavioral evaluations for subagent routing ( #23272 )
...
Co-authored-by: Samee Zahid <sameez@google.com >
2026-03-24 01:19:21 +00:00
f784e192fa
eval(save_memory): add multi-turn interactive evals for memoryManager ( #23572 )
2026-03-23 22:58:55 +00:00
db14cdf92b
feat(skills): add behavioral-evals skill with fixing and promoting guides ( #23349 )
2026-03-23 21:06:43 +00:00
0df9498674
fix(core): refine CliHelpAgent description for better delegation ( #23310 )
2026-03-21 06:24:37 +00:00
d3766875f8
fix(evals): remove tool restrictions and add compile-time guards ( #23312 )
2026-03-21 03:45:33 +00:00
28935d1e6b
Retry evals on API error. ( #23322 )
2026-03-21 02:52:19 +00:00
fbb17ebf58
Disabling failing test while investigating ( #23311 )
2026-03-20 22:52:35 +00:00
1725ec346b
feat(plan): support plan mode in non-interactive mode ( #22670 )
2026-03-18 20:00:26 +00:00
fe8d93c75a
Promote stable tests. ( #22253 )
2026-03-13 21:32:00 +00:00
263b8cd3b3
fix(plan): Fix AskUser evals ( #22074 )
2026-03-13 13:30:19 +00:00
2dd037682c
Add behavioral evals for tracker ( #20069 )
2026-03-10 18:51:54 +00:00
4669148a4c
feat(core): add concurrency safety guidance for subagent delegation ( #17753 ) ( #21278 )
2026-03-06 18:09:45 +00:00
fe332bbef7
feat(evals): add behavioral evals for ask_user tool ( #20620 )
2026-03-03 17:51:15 +00:00
25f59a0099
Add some dos and don'ts to behavioral evals README. ( #20629 )
2026-03-02 23:14:00 +00:00
05ef2eb362
Promote stable tests to CI blocking. ( #20581 )
2026-02-27 21:08:12 +00:00
b2b6092c24
Add slash command for promoting behavioral evals to CI blocking ( #20575 )
...
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
2026-02-27 19:11:30 +00:00
514d431049
Demote unreliable test. ( #20571 )
2026-02-27 16:48:46 +00:00
611d934829
feat(core): Enable generalist agent ( #19665 )
2026-02-26 16:38:49 +00:00
39938000a9
feat(core): rename grep_search include parameter to include_pattern ( #20328 )
2026-02-26 04:16:21 +00:00
56c8d7e985
Stabilize tests. ( #20095 )
2026-02-24 00:01:39 +00:00
27b7fc04de
Search updates ( #19482 )
...
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
2026-02-20 17:54:28 +00:00
87f5dd15d6
feat(core): experimental in-progress steering hints (2 of 2) ( #19307 )
2026-02-18 22:05:50 +00:00
8f6a711a3a
fix(core): clarify plan mode constraints and exit mechanism ( #19438 )
2026-02-18 20:09:59 +00:00
8a8826654c
Disable failing eval test ( #19455 )
2026-02-18 19:27:21 +00:00
f1aa38b258
test(evals): add behavioral tests for tool output masking ( #19172 )
2026-02-18 05:07:25 +00:00
ce84b3cb5f
Use ranged reads and limited searches and fuzzy editing improvements ( #19240 )
2026-02-17 23:54:08 +00:00
55c628e967
feat(core): experimental in-progress steering hints (1 of 3) ( #19008 )
2026-02-17 22:59:33 +00:00
6eec9f3350
fix(core): Encourage non-interactive flags for scaffolding commands ( #18804 )
2026-02-15 20:26:59 +00:00
Mahima Shanware