Commit Graph

817 Commits

Author SHA1 Message Date
Abhi
331ae7dbdf feat: Add enableSubagents configuration and wire up subagent registration (#9988) 2025-10-01 20:54:00 +00:00
Jerop Kipruto
5b16771567 feat(telemetry): add OpenTelemetry GenAI semantic convention metrics (#10343) 2025-10-01 20:24:55 +00:00
Silvio Junior
7e493f4a9e Codebase Investigator: Separate initial query from system prompt and apply templateStrings in query and initialMessages (#10282) 2025-10-01 20:21:01 +00:00
Christie Warwick (Wilson)
5c6f006634 Refactor metrics definitions to be easily understandable (#10215) 2025-10-01 13:33:47 +00:00
Allen Hutchison
f207ea94d6 fix(memory): ignore @ inside code blocks (#10201) 2025-10-01 01:24:33 +00:00
Abhi
794d92a79d refactor(agents): Introduce Declarative Agent Framework (#9778) 2025-09-30 21:00:54 +00:00
Gal Zahavi
6695c32aa2 fix(shell): improve shell output presentation and usability (#8837) 2025-09-30 20:44:58 +00:00
Vikram Pasupathy
953935d67c Fix a cache collision bug in the llm edit fixer (#9542)
Co-authored-by: anthony bushong <agmsb@users.noreply.github.com>
2025-09-30 16:58:32 +00:00
Victor May
ec08129fba Regex Search/Replace for Smart Edit Tool (#10178) 2025-09-30 16:06:03 +00:00
Dmytro Salnyk
1d24f95a3f fix(telemetry): Improve Cloud Shell surface type detection for telemetry purposes (#10162) 2025-09-30 10:35:34 +00:00
Adam Weidman
0c3fcb7030 chore(mocktools): final step in unify mock tool definitions (#9166) 2025-09-29 20:43:06 +00:00
Sandy Tao
ac4a79223a feat(core): Add content-based retries for JSON generation (#9264) 2025-09-29 19:27:15 +00:00
Sandy Tao
8a2c2dc73f feat(core): Enable tool output truncation by default (#9983) 2025-09-29 16:30:37 +00:00
anthony bushong
1bd75f060d fix(core): auto-correct file paths in smart edit where possible (x-platform) (#10018) 2025-09-27 23:16:51 +00:00
Sandy Tao
ffcd996366 feat(core): Use lastPromptTokenCount to determine if we need to compress (#10000) 2025-09-27 18:56:10 +00:00
Tommaso Sciortino
93694c6a65 Make compression algo slightly more aggressive (#10024) 2025-09-27 12:15:53 +00:00
Adam Weidman
db51e3f4cd feat(iap support): Add service account impersonation provider to MCPServers to support IAP on Cloud Run (#8505)
Co-authored-by: Bryan Morgan <bryanmorgan@google.com>
2025-09-27 08:12:24 +00:00
Gal Zahavi
eb1a6a6091 Revert "fix(core): auto-correct file paths in smart edit where possible" (#10009) 2025-09-27 01:05:48 +00:00
Gal Zahavi
24c15b9d43 Revert "Make --allowed-tools work in non-interactive mode" (#10006) 2025-09-26 16:22:02 -07:00
anthony bushong
0d22b22c82 fix(core): auto-correct file paths in smart edit where possible (#9526) 2025-09-26 20:05:15 +00:00
owenofbrien
e909993dd1 Added warning to avoid command substitution in run_shell_command tool… (#9934)
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
2025-09-26 18:16:05 +00:00
Abhi
3d7cb3fb8a refactor(core): Extract file filtering constants from Config to break circular dependency (#9974) 2025-09-26 17:37:00 +00:00
mistergarrison
e8a065cb9f Make --allowed-tools work in non-interactive mode (#9114)
Co-authored-by: google-labs-jules[bot] <161369871+google-labs-jules[bot]@users.noreply.github.com>
2025-09-26 17:03:35 +00:00
geoffdowns
11c995e9fa Stop checking MCP tool schemas for type definitions (#9574)
Co-authored-by: Gal Zahavi <38544478+galz10@users.noreply.github.com>
Co-authored-by: anthony bushong <agmsb@users.noreply.github.com>
Co-authored-by: Abhi <43648792+abhipatel12@users.noreply.github.com>
Co-authored-by: Sandy Tao <sandytao520@icloud.com>
Co-authored-by: Jacob MacDonald <jakemac@google.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Co-authored-by: christine betts <chrstn@uw.edu>
Co-authored-by: Adam Weidman <65992621+adamfweidman@users.noreply.github.com>
Co-authored-by: matt korwel <matt.korwel@gmail.com>
Co-authored-by: Shreya Keshive <skeshive@gmail.com>
Co-authored-by: HugoMurillo <mhugorodrigo@google.com>
Co-authored-by: Shreya Keshive <shreyakeshive@google.com>
Co-authored-by: Miguel Solorio <miguelsolorio@google.com>
Co-authored-by: Christie Warwick (Wilson) <bobcatfish@gmail.com>
Co-authored-by: shrutip90 <shruti.p90@gmail.com>
2025-09-26 04:42:21 +00:00
Gal Zahavi
6535b71c35 fix(prompt): Prevent model from reverting successful changes (#9253) 2025-09-25 20:54:44 +00:00
Sandy Tao
e209724789 fix(core): Improve API error retry logic (#9763) 2025-09-25 17:57:00 +00:00
anthony bushong
4caaa2a8e8 fix(core): ensure retry sets defaults for nullish values passed into options (#9540)
Co-authored-by: Abhi <43648792+abhipatel12@users.noreply.github.com>
2025-09-25 17:33:00 +00:00
Adrian Arribas
f80eb71068 [Part 3/6] feat(telemetry): enhance metrics with performance monitoring APIs (#8113) 2025-09-25 16:12:46 +00:00
Shardul Natu
135d3401cd add(telemetry): Add character-level edit metrics to Concord (#9145)
Co-authored-by: Shnatu <snatu@google.com>
Co-authored-by: owenofbrien <86964623+owenofbrien@users.noreply.github.com>
2025-09-25 16:09:56 +00:00
anthony bushong
275a12fd45 fix(core): set default maxAttempts in baseLLMClient (#9533) 2025-09-25 01:18:26 +00:00
anthony bushong
8abe7e151c fix(core): plumb max attempts for retry to generate options in baseLLMClient (#9518) 2025-09-24 22:25:15 +00:00
anthony bushong
05c962af1f fix(core): update edit tool error type during llm judgements (#9510) 2025-09-24 22:19:19 +00:00
jleong-stripe
e0ba7e4ffb For dynamic client registration - use registration endpoint in config if available instead of performing OAuth discovery again (#9231) 2025-09-24 20:53:42 +00:00
Abhi
22740ddceb refactor(core): Extract thought parsing logic into a dedicated utility (#9503) 2025-09-24 19:38:36 +00:00
Sandy Tao
ad59be0c81 fix(core): Fix unable to cancel edit tool (#9299) 2025-09-24 19:16:00 +00:00
fuyou
66c2184fe5 feat: Add AbortSignal support for retry logic and tool execution (#9196)
Co-authored-by: Sandy Tao <sandytao520@icloud.com>
2025-09-24 19:10:55 +00:00
geoffdowns
4f49341ce9 relax JSON schema validation (#9332)
Co-authored-by: Tommaso Sciortino <sciortino@gmail.com>
2025-09-24 16:55:32 +00:00
Victor May
4bd4cd697b Log Model Slash Commands (#9250) 2025-09-23 22:06:03 +00:00
Sandy Tao
28c3901513 feat(logging): Add model name to content retry events (#9266) 2025-09-23 21:34:17 +00:00
Abhi
c12b87ed03 fix(routing): Round latency to integer for telemetry (#9242) 2025-09-23 21:29:12 +00:00
shishu314
98461ff667 metrics(extension) - Add logging for disable extension (#9238)
Co-authored-by: Shi Shu <shii@google.com>
2025-09-23 18:37:35 +00:00
Abhi
c93eed6384 fix(core): Compression was broken when routing enabled (#9183) 2025-09-23 04:44:07 +00:00
Sandy Tao
712dc245ac fix(quality): Retry invalid stream only 1 time (#9169) 2025-09-23 00:15:55 +00:00
Sandy Tao
4cdf9207f3 feat(core): Log invalid stream type (#9168) 2025-09-22 23:39:44 +00:00
Allen Hutchison
6328ca512f Fix: Only log message bus debug info when feature is enabled (#9156) 2025-09-22 23:08:48 +00:00
Sandy Tao
9c4d1594ac Fix(quality): Refine Stream Validation Logic (#9150) 2025-09-22 21:40:09 +00:00
Allen Hutchison
bcc4d81d19 feat: add message bus integration for tool confirmation (#8938) 2025-09-22 19:03:20 +00:00
Jerop Kipruto
c0c7ad10ca feat(config): Support telemetry configuration via environment variables (#9113) 2025-09-22 18:23:06 +00:00
shishu314
10392ad344 Telemtry(extensions) - Add logging for extension events (#8339)
Co-authored-by: Shi Shu <shii@google.com>
2025-09-22 16:55:43 +00:00
Jacob MacDonald
34c14b7dd2 Store installed release version as a part of install metadata (#9122) 2025-09-22 16:42:35 +00:00