When Claude Code strips thinking signatures it doesn't recognize,
the proxy would drop unsigned thinking blocks, causing the error
"Expected thinking but found text". This fix detects unsigned
thinking blocks and triggers recovery to close the tool loop.
Co-Authored-By: Claude <noreply@anthropic.com>
The tier detection was incorrectly showing Pro accounts as "free" due to:
1. paidTier field being flaky/missing from some API responses
2. standard-tier not being recognized as a Pro tier
3. No fallback to allowedTiers when currentTier is missing
Changes:
- Add parseTierId() helper to centralize tier ID parsing
- Recognize "standard-tier" as Pro (Gemini Code Assist paid tier)
- Add fallback chain: paidTier > currentTier > allowedTiers
- Return "unknown" instead of incorrectly defaulting to "free"
Fixes#121
Co-Authored-By: Claude <noreply@anthropic.com>
When multiple tool_results contain images, the inlineData parts were
being interleaved between functionResponse parts, breaking Claude's API
requirement that functionResponse parts be consecutive.
Co-Authored-By: Claude <noreply@anthropic.com>
Replace gpt-tokenizer with model-specific official tokenizers:
- Claude models: @anthropic-ai/tokenizer (official Anthropic tokenizer)
- Gemini models: @lenml/tokenizer-gemini (GemmaTokenizer)
Changes:
- Add @anthropic-ai/tokenizer and @lenml/tokenizer-gemini dependencies
- Remove gpt-tokenizer dependency
- Update count-tokens.js with model-aware tokenization
- Use getModelFamily() to select appropriate tokenizer
- Lazy-load Gemini tokenizer (138MB) on first use
- Default to local estimation for all content types (no API calls)
Tested with all supported models:
- claude-sonnet-4-5, claude-opus-4-5-thinking, claude-sonnet-4-5-thinking
- gemini-3-flash, gemini-3-pro-low, gemini-3-pro-high
- Add ensureInitialized() call before count_tokens handler
- Use hybrid approach: local estimation for text, API for images/docs
- This prevents "No accounts available" error on first request
Switch from local estimation (gpt-tokenizer) to API-based counting
via Google Cloud Code API for accurate token counts. Falls back to
local estimation if API call fails.
Add Anthropic-compatible token counting endpoint using hybrid approach:
- Local estimation with gpt-tokenizer for text content (~95% accuracy)
- API-based counting for complex content (images, documents)
- Automatic fallback to local estimation on API errors
This resolves warnings in LiteLLM and other clients that rely on
pre-request token counting.
- Move TEST_MODELS and DEFAULT_PRESETS to src/constants.js as single source of truth
- Update test-models.cjs helper to use dynamic import from constants
- Make getTestModels() and getModels() async functions
- Update all test files to await async model config loading
- Remove duplicate THINKING_MODELS and getThinkingModels() from test helper
- Make thinking tests more lenient for Gemini (doesn't always produce thinking blocks)
Co-Authored-By: Claude <noreply@anthropic.com>
- Cooldown now caps API-provided reset times instead of being a fallback
- Fixed misleading UI descriptions for cooldown settings
- Removed unused cooldownDurationMs from settings object
- Updated default fallback values in frontend to 10s
Co-Authored-By: Claude <noreply@anthropic.com>
- Add separate LOAD_CODE_ASSIST_ENDPOINTS (prod first) and
LOAD_CODE_ASSIST_HEADERS (google-api-nodejs-client User-Agent)
- Add duetProject to metadata for project discovery
- Silent fallback when API returns success but no project
(matches opencode-antigravity-auth behavior)
- Only warn when all endpoints fail with actual errors
Fixes#114, addresses discussion #113
Co-Authored-By: Claude <noreply@anthropic.com>
- Add backend storage logic in `src/utils/claude-config.js` to save/load/delete presets
- Add API endpoints (`GET`, `POST`, `DELETE`) for presets in `src/webui/index.js`
- Update `public/views/settings.html` with new Presets UI card and modals
- Update `public/js/components/claude-config.js` with auto-load logic and unsaved changes protection
- Add translations (EN/ZH) for new UI elements in `public/js/store.js`
- Add integration tests in `tests/frontend/test-frontend-settings.cjs`
- Update compiled CSS
Co-Authored-By: Claude <noreply@anthropic.com>
When a tool call is made, stopReason is set to 'tool_use'. However, when
finishReason: STOP arrives later, it was overwriting stopReason back to
'end_turn', breaking multi-turn tool conversations in clients like OpenCode.
Fix: Initialize stopReason to null and only set it from finishReason if
not already set. This ensures tool_use is preserved once detected.
Fixes#96
Co-Authored-By: Claude <noreply@anthropic.com>
When all accounts fail with HTTP 500/503 errors (e.g., Google API returning
'Unknown Error' for Claude models on large conversations), the proxy now
attempts to use a fallback model if --fallback is enabled.
This enables graceful degradation when:
- All accounts are exhausted due to 5xx errors (not just rate limits)
- Claude models fail on very large conversations
- The API has temporary issues with specific models
The fallback uses the existing MODEL_FALLBACK_MAP configuration:
- claude-opus-4-5-thinking -> gemini-3-pro-high
- claude-sonnet-4-5-thinking -> gemini-3-flash
Relates to #88
- Update CLAUDE.md with comprehensive WebUI architecture and API documentation
- Improve settings UI with searchable model dropdowns and visual family indicators
- Migrate usage statistics persistence to user config directory with auto-migration
- Refactor server request handling and fix model suffix logic
The /compact command was failing with 'Proto field is not repeating,
cannot start list' error for Claude models because tool schemas were
sent with lowercase JSON Schema types (array, object, string) but
Google's Cloud Code API expects uppercase protobuf types (ARRAY,
OBJECT, STRING).
Changes:
- Add toGoogleType() function to convert JSON Schema types to Google format
- Add Phase 5 to cleanSchemaForGemini() for type conversion
- Apply cleanSchemaForGemini() for ALL models (not just Gemini) since
all requests go through Cloud Code API which validates schema format
- Add comprehensive test suite with 10 tests covering nested arrays,
complex schemas, and real-world Claude Code tool scenarios
Fixes#82
- Add [ignore] tags around system instruction to prevent model from
identifying as "Antigravity" when asked "Who are you?"
- Replace full system instruction with minimal version used by
CLIProxyAPI/gcli2api to reduce token usage and improve response quality
Fixes#76
Co-Authored-By: Claude <noreply@anthropic.com>
Fixes issue #71 - 'No accounts available' error when API returns 429
The Google Cloud Code API can return 429 RESOURCE_EXHAUSTED errors even
when accounts have quota available due to:
- Temporary API load/throttling
- Per-minute request limits (not per-day quota)
- Transient backend issues
This fix adds:
1. A 500ms buffer after waiting for rate limits to expire
2. Optimistic rate limit reset when all accounts appear stuck
3. Retry logic that clears rate limits and tries again
The fix works in conjunction with the server-level optimistic reset
that already exists, providing multiple layers of protection against
false 'No accounts available' errors.
Co-Authored-By: Claude <noreply@anthropic.com>
## Dashboard Enhancements
- Add Request Volume trend chart with Chart.js line graph
- Support Family/Model display modes for aggregation levels
- Show Total/Today/1H usage statistics
- Hierarchical filter dropdown with Smart select (Top 5 by 24h usage)
- Persist chart preferences to localStorage
- Improve account health detection logic
- Core models (sonnet/opus/pro/flash) require >5% quota to be healthy
- Dynamic quota ring chart supporting any model family
- Unify table styles with standard-table class
## Global Style Refactoring
- Add CSS variable system for theming
- Space color scale (950/900/850/800/border)
- Neon accent colors (purple/green/cyan/yellow/red)
- Text hierarchy (main/dim/muted/bright)
- Chart palette (16 colors)
- Add unified component classes
- .view-container for consistent page layouts
- .section-header/.section-title/.section-desc
- .standard-table for table styling
- Update scrollbar, nav-item, progress-bar to use theme variables
## Settings Module Extensions
- Add model mapping column in Models tab
- Enhance model selectors with family color indicators
- Support horizontal scroll for tabs on narrow screens
- Add defaultCooldownMs and maxWaitBeforeErrorMs config options
## New Module
- Add src/modules/usage-stats.js for request tracking
- Track /v1/messages and /v1/chat/completions endpoints
- Hierarchical storage: { hour: { family: { model: count } } }
- Auto-save every minute, 30-day retention
- GET /api/stats/history endpoint for dashboard chart
## Backend Changes
- Add direct account manipulation helpers (bypass AccountManager)
- Add POST /api/config/password endpoint for WebUI password change
- Auto-reload AccountManager after account operations
- Use CSS variables in OAuth callback pages
## Other
- Update .gitignore for runtime data directory
- Add i18n keys for new UI elements (EN/zh_CN)
Co-Authored-By: Claude <noreply@anthropic.com>
Issues found by Claude Opus 4.5 + Gemini 3 Pro:
HIGH PRIORITY FIXES:
- Mark account rate-limited when 429 occurs during retry (was losing resetMs)
- Add exponential backoff between retries (500ms, 1000ms, 2000ms)
- Fix 5xx handling: don't pass error response to streamer, refetch instead
- Use recognizable error messages (429/401) for isRateLimitError/isAuthError
MEDIUM PRIORITY FIXES:
- Refactor while loop to for loop for clearer retry semantics
- Simplify logic flow with early returns
Code review by: Claude Opus 4.5 + Gemini 3 Pro Preview
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Move MAX_EMPTY_RESPONSE_RETRIES to constants.js for consistency
- Handle 429/401/5xx errors properly during retry fetch
- Use proper message ID format (crypto.randomBytes) instead of Date.now()
- Add crypto import for UUID generation
Code review by: Gemini 3 Pro Preview + Claude Opus 4.5
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
When Claude Code sends requests with large thinking_budget values,
the model may spend all tokens on "thinking" and return empty responses,
causing Claude Code to stop mid-conversation.
This commit adds a retry mechanism that:
- Throws EmptyResponseError instead of emitting fake message on empty response
- Retries up to 2 times before giving up
- Emits fallback message only after all retries are exhausted
Changes:
- src/errors.js: Added EmptyResponseError class and isEmptyResponseError()
- src/cloudcode/sse-streamer.js: Throw error instead of yielding fake message
- src/cloudcode/streaming-handler.js: Added retry loop with fallback
Tested for 6+ hours with 1,884 API requests and 88% recovery rate
on empty responses.
Fixes#61🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Ensure proper encoding for international characters in error and success pages
by specifying charset=utf-8 in Content-Type headers and adding meta charset tags.
Replace generic Error instances with NativeModuleError in loadDatabaseModule
to provide more structured error information, including rebuild status and
restart requirements. Update getAuthStatus to re-throw NativeModuleError
instances without wrapping.
Add a new error class to handle native module errors, including version mismatches and rebuild requirements. This supports the auto-rebuild functionality by providing structured error information for rebuild success and restart needs.
