Commit graph

25 commits

Author SHA1 Message Date
Mario Zechner
ed9255d3bd ai: update Antigravity system instruction to compact version, fix flaky tool test
- Replace verbose ANTIGRAVITY_SYSTEM_INSTRUCTION with compact version from CLIProxyAPI
- Replace bridgePrompt override with [ignore] wrapper pattern
- Switch Antigravity Gemini test model from gemini-3-flash to gemini-3-pro-high
- Rename calculator tool to math_operation (gemini-3-pro ignores schema for 'calculator')

closes #1415
2026-02-08 23:37:09 +01:00
Mario Zechner
1caadb2e2a fix(ai): use parametersJsonSchema for Google tool declarations, revert Antigravity opus model
- Use parametersJsonSchema instead of parameters for Gemini tool declarations
  to support full JSON Schema (anyOf, oneOf, const, etc.)
- Keep legacy parameters field for Claude models on Cloud Code Assist, where
  the API translates parameters into Anthropic's input_schema
- Revert claude-opus-4-6-thinking back to claude-opus-4-5-thinking (model
  doesn't exist on the Antigravity endpoint)

fixes #1398
2026-02-08 15:53:17 +01:00
Mario Zechner
e9ca0be769 feat(ai): add PI_AI_ANTIGRAVITY_VERSION env var override
Allows users to override the Antigravity User-Agent version when Google
updates their version requirements, avoiding the need to wait for a
package release.

Fixes #1129
2026-02-01 09:32:10 +01:00
Mario Zechner
030a61d88c feat: add maxDelayMs setting to cap server-requested retry delays
When a provider (e.g., Google Gemini CLI) requests a retry delay longer
than maxDelayMs (default: 60s), the request fails immediately with an
informative error instead of waiting silently for hours.

The error is then handled by agent-level auto-retry, which shows the
delay to the user and allows aborting with Escape.

- Add maxRetryDelayMs to StreamOptions (packages/ai)
- Add maxRetryDelayMs to AgentOptions (packages/agent)
- Add retry.maxDelayMs to settings (packages/coding-agent)
- Update _isRetryableError to match 'retry delay' errors

fixes #1123
2026-02-01 00:50:41 +01:00
Mario Zechner
52532c7c00 fix(ai): update Antigravity User-Agent to 1.15.8
fixes #1079
2026-01-29 23:10:49 +01:00
Mario Zechner
a6d878e804 fix(ai): default tool call arguments to empty object for Google providers
When Google providers return tool calls without an args field (common for
no-argument tools), the arguments field was undefined. This breaks
subsequent API calls that require tool_use.input to be present.

Now defaults to {} when args is missing.

Related: clawdbot/clawdbot#1509
2026-01-25 03:18:02 +01:00
Mario Zechner
c725135a76 refactor(ai): register api providers 2026-01-24 23:15:11 +01:00
Mario Zechner
d2be6486a4 feat(ai): add headers option to StreamOptions for custom HTTP headers
- Added headers field to base StreamOptions interface
- Updated all providers to merge options.headers with defaults
- Forward headers and onPayload through streamSimple/completeSimple
- Bedrock not supported (uses AWS SDK auth)
2026-01-20 01:08:31 +01:00
Mario Zechner
2c7c23b865 fix(ai): normalize tool call ids and handoff tests fixes #821 2026-01-19 00:10:49 +01:00
Ahmed Kamal
ff15414258
Improve Gemini CLI provider retries and headers (#670)
Improve Gemini CLI provider retries and headers

- Add Antigravity endpoint fallback (tries daily sandbox then prod when baseUrl is unset)
- Parse retry delays from headers (Retry-After, x-ratelimit-reset, x-ratelimit-reset-after) before body parsing
- Derive stable sessionId from first user message for cache affinity
- Retry empty SSE streams with backoff without duplicate start/done events
- Add anthropic-beta header for Claude thinking models only
2026-01-13 01:04:53 +01:00
Danila Poyarkov
9e4ae98358
Improve Google Cloud Code Assist error handling (#665)
* Improve Cloud Code Assist error messages

- Extract just the message from verbose JSON error responses
- Extract cause from generic 'fetch failed' errors for better diagnostics

* Make 'other side closed' network error retryable

* Make 'other side closed' network error retryable
2026-01-13 00:41:20 +01:00
theBucky
a315cfe813 fix(ai): complete textSignature round-trip for Google providers
- Store thoughtSignature on text blocks during streaming (all 3 providers)
- Replay textSignature as thoughtSignature in convertMessages
- Remove redundant conditional since retainThoughtSignature handles undefined

Per Google docs, text part signatures are optional but recommended for
high-quality reasoning in multi-turn conversations.
2026-01-11 19:25:38 +01:00
theBucky
4f757fbe23 fix(ai): correct Google thinking detection and remove unsupported id fields
- isThinkingPart now only checks thought === true, not thoughtSignature
- thoughtSignature is for context replay and can appear on any part type
- Store thoughtSignature on text blocks as textSignature for proper replay
- Remove id from functionCall/functionResponse (unsupported by Vertex/Cloud Code Assist)

Refs: https://ai.google.dev/gemini-api/docs/thought-signatures
Co-authored-by: Amp <amp@ampcode.com>
2026-01-11 19:25:38 +01:00
Mario Zechner
ef7c52ffa1 chore: fix template literal lint, update AGENTS.md to require fixing all check output 2026-01-08 23:44:26 +01:00
Mario Zechner
aa89080ea0 fix(ai): add bridge prompt to override Antigravity behavior with Pi defaults 2026-01-08 20:20:24 +01:00
Mario Zechner
31f155d7db
Merge pull request #571 from ben-vargas/fix-antigravity-patch
fix(ai): align antigravity request payload
2026-01-08 20:00:40 +01:00
Ben Vargas
74476be61d fix(ai): align antigravity request payload 2026-01-08 10:00:44 -07:00
Thomas Mustier
a65da1c14b fix: ESC key not interrupting during Working... state
Three related fixes:

1. google-gemini-cli: Handle abort signal in stream reading loop
   - Add abort event listener to cancel reader immediately when signal fires
   - Fix AbortError detection in retry catch block (fetch throws AbortError,
     not our custom message)
   - Swallow reader.cancel() rejection to avoid unhandled promise

2. agent-session: Fix retry attempt counter showing 0 on cancel
   - abortRetry() was resetting _retryAttempt before the catch block could
     read it for the error message

3. interactive-mode: Restore main escape handler on agent_start
   - When auto-retry starts, onEscape is replaced with retry-specific handler
   - auto_retry_end (which restores it) fires on turn_end, after streaming begins
   - Now restore immediately on agent_start if retry handler is still active

Amended: suppress reader.cancel() rejection on abort.
2026-01-08 12:35:34 +00:00
Ahmed Kamal
e42e9e6305 fix(ai): classify Google thoughtSignature as thinking
Google streaming may emit thoughtSignature without thought=true (including empty-text signature-only parts). Treat non-empty thoughtSignature as thinking to avoid leaking reasoning into normal text and retain signature across streaming deltas. Add unit test coverage.
2026-01-06 20:47:19 +02:00
Mario Zechner
fd35d9188c fix(ai): add retry with server-provided delay for Gemini CLI rate limits, fixes #370 2026-01-02 10:45:59 +01:00
Mario Zechner
ecd240f636 Define own GoogleThinkingLevel type instead of importing from @google/genai
- Add GoogleThinkingLevel type mirroring Google's ThinkingLevel enum
- Update GoogleGeminiCliOptions and GoogleOptions to use our type
- Cast to any when assigning to Google SDK's ThinkingConfig
2025-12-30 22:42:25 +01:00
Mario Zechner
329b3a0a36 Merge PR #264: Add Gemini 3 preview models to google-gemini-cli provider 2025-12-21 20:31:19 +01:00
Luke Foster
ee9b498380 Add Gemini 3 preview models to google-gemini-cli provider
- Add gemini-3-pro-preview and gemini-3-flash-preview to Cloud Code Assist
- Handle thinkingLevel config for Gemini 3 (vs thinkingBudget for Gemini 2.x)
- Gemini 3 Pro: LOW/HIGH levels only
- Gemini 3 Flash: all four levels (MINIMAL/LOW/MEDIUM/HIGH)
2025-12-20 22:10:47 -06:00
Mario Zechner
2a0283ecfd Fix token counts in google-gemini-cli.ts for anthropic models. 2025-12-20 21:34:18 +01:00
Mario Zechner
c359023c3f Add Google Gemini CLI and Antigravity OAuth providers
- Add google-gemini-cli provider: free Gemini 2.0/2.5 via Cloud Code Assist
- Add google-antigravity provider: free Gemini 3, Claude, GPT-OSS via sandbox
- Move OAuth infrastructure from coding-agent to ai package
- Fix thinking signature handling for cross-model handoff
- Fix OpenAI message ID length limit (max 64 chars)
- Add GitHub Copilot overflow pattern detection
- Add OAuth provider tests for context overflow and streaming
2025-12-20 21:34:18 +01:00
Renamed from packages/ai/src/providers/google-cloud-code-assist.ts (Browse further)