- Add archon_configured_repositories table migration with production-ready sandbox type constraints
- Implement SupabaseWorkOrderRepository for CRUD operations with comprehensive error handling
- Add defensive validation in _row_to_model with detailed logging for invalid enum values
- Implement granular exception handling (409 duplicates, 422 validation, 502 GitHub API errors)
- Document async/await pattern for interface consistency across repository implementations
- Add Supabase health check to verify table existence
- Expand test coverage from 10 to 17 tests with error handling and edge case validation
- Add supabase dependency to agent-work-orders group
- Enable ENABLE_AGENT_WORK_ORDERS flag in docker-compose for production deployment
Add ENABLE_AGENT_WORK_ORDERS configuration flag to allow disabling the agent work orders microservice. Service discovery now gracefully handles unavailable services, and health checks return appropriate status when feature is disabled.
Generated with Claude Code
Co-Authored-By: Claude <noreply@anthropic.com>
- Move sse-starlette from base dependencies to agent-work-orders group
- Keep structlog in agent-work-orders group (already there)
- Update lockfile accordingly
- Change from fixed backend/frontend ports to 10-port ranges per work order
- Support 20 concurrent work orders (200 ports: 9000-9199)
- Add port availability checking with flexible allocation
- Make git_worktree default sandbox type
- Standardize API routes with /api/ prefix
- Add comprehensive port allocation tests
- Update environment file generation with PORT_0-PORT_9 variables
- Maintain backward compatibility with BACKEND_PORT/FRONTEND_PORT aliases
Aligns test expectations with the llms.txt specification which uses
'pages' rather than 'files' terminology. The implementation correctly
uses "llms_txt_with_linked_pages" - this updates the test to match.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
The tldextract package was missing from the 'all' dependency group,
causing CI test failures. It was already in the 'server' group but
needed in 'all' for running unit tests in CI/CD.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Fix test mocks to use requests.Session for _check_url_exists
- Add url parameter to create_mock_response to prevent MagicMock issues
- Update all test scenarios to mock both requests.get and session.get
- Remove redundant UNSAFE_PROTOCOLS check in URL validation
- Fix test assertions to match new priority order (llms.txt > llms-full.txt)
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
## Backend Improvements
### Discovery Service
- Fix SSRF protection: Use requests.Session() for max_redirects parameter
- Add comprehensive IP validation (_is_safe_ip, _resolve_and_validate_hostname)
- Add hostname DNS resolution validation before requests
- Fix llms.txt link following to crawl ALL same-domain pages (not just llms.txt files)
- Remove unused file variants: llms.md, llms.markdown, sitemap_index.xml, sitemap-index.xml
- Optimize DISCOVERY_PRIORITY based on real-world usage research
- Update priority: llms.txt > llms-full.txt > sitemap.xml > robots.txt
### URL Handler
- Fix .well-known path to be case-sensitive per RFC 8615
- Remove llms.md, llms.markdown, llms.mdx from variant detection
- Simplify link collection patterns to only .txt files (most common)
- Update llms_variants list to only include spec-compliant files
### Crawling Service
- Add tldextract for proper root domain extraction (handles .co.uk, .com.au, etc.)
- Replace naive domain extraction with robust get_root_domain() function
- Add tldextract>=5.0.0 to dependencies
## Frontend Improvements
### Type Safety
- Extend ActiveOperation type with discovery fields (discovered_file, discovered_file_type, linked_files)
- Remove all type casting (operation as any) from CrawlingProgress component
- Add proper TypeScript types for discovery information
### Security
- Create URL validation utility (urlValidation.ts)
- Only render clickable links for validated HTTP/HTTPS URLs
- Reject unsafe protocols (javascript:, data:, vbscript:, file:)
- Display invalid URLs as plain text instead of links
## Testing
- Update test mocks to include history and url attributes for redirect checking
- Fix .well-known case sensitivity tests (must be lowercase per RFC 8615)
- Update discovery priority tests to match new order
- Remove tests for deprecated file variants
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
GitHub's YAML templates (.yml) don't support URL parameter pre-filling, but
Markdown templates (.md) do. This adds a structured bug report template that
allows the automated bug reporter to pre-fill all user-submitted data.
Changes:
- Create .github/ISSUE_TEMPLATE/auto_bug_report.md template
- Update bug_report_api.py to use template=auto_bug_report.md parameter
- Update tests to verify template parameter is included in URL
- Add explanatory comments about YAML vs Markdown template differences
Benefits:
- Users see a structured bug report template (not generic issue form)
- All bug report data is pre-filled from the UI form
- Template provides consistent formatting and organization
- Better UX than generic issue creation
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
GitHub's issue creation URL does not support the 'template' parameter for
pre-filling fields. When a template is specified, GitHub ignores other URL
parameters like title and body, preventing user-submitted data from being
pre-filled in the issue form.
Changes:
- Remove 'template=bug_report.yml' parameter (non-existent template)
- Remove 'labels' parameter (not supported via URL)
- Keep only 'title' and 'body' parameters for proper pre-filling
- Add explanatory comment about GitHub's URL parameter limitations
- Update tests to verify URL structure (no template parameter)
Now when users click "Report Bug", the GitHub issue form will be properly
pre-filled with their title and detailed bug report information.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Fixes#802
The bug report feature was redirecting users to the old repository URL
(dynamous-community/Archon-V2-Alpha) instead of the current repository
(coleam00/Archon). This occurred because hardcoded default values in the
bug report API were not updated during the Alpha-to-Beta rebranding.
Changes:
- Import GITHUB_REPO_OWNER and GITHUB_REPO_NAME from version.py
- Update GitHubService.__init__() to construct default from constants
- Update health check endpoint to use same centralized default
- Add comprehensive integration tests for bug report URL generation
- Document repository configuration in CLAUDE.md
The fix ensures single source of truth for repository information and
maintains backward compatibility with GITHUB_REPO environment variable override.
All tests pass (7/7) validating correct repository URL usage.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Replace dot-based file detection with explicit extension checking
in discovery service to correctly handle versioned directories
like /docs.v2
- Add comprehensive validation for start_progress and end_progress
parameters in crawl_markdown_file to ensure they are valid
numeric values in range [0, 100] with start < end
- Validation runs before any async work or progress reporting begins
- Clear error messages indicate which parameter is invalid and why
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Implements complete llms.txt link following functionality that crawls
linked llms.txt files on the same domain/subdomain, along with critical
bug fixes for discovery priority and variant detection.
Backend Core Functionality:
- Add _is_same_domain_or_subdomain method for subdomain matching
- Fix is_llms_variant to detect .txt files in /llms/ directories
- Implement llms.txt link extraction and following logic
- Add two-phase discovery: prioritize ALL llms.txt before sitemaps
- Enhanced progress reporting with discovery metadata
Critical Bug Fixes:
- Discovery priority: Fixed sitemap.xml being found before llms.txt
- is_llms_variant: Now matches /llms/guides.txt, /llms/swift.txt, etc.
- These were blocking bugs preventing link following from working
Frontend UI:
- Add discovery and linked files display to CrawlingProgress component
- Update progress types to include discoveredFile, linkedFiles fields
- Add new crawl types: llms_txt_with_linked_files, discovery_*
- Add "discovery" to ProgressStatus enum and active statuses
Testing:
- 8 subdomain matching unit tests (test_crawling_service_subdomain.py)
- 7 integration tests for link following (test_llms_txt_link_following.py)
- All 15 tests passing
- Validated against real Supabase llms.txt structure (1 main + 8 linked)
Files Modified:
Backend:
- crawling_service.py: Core link following logic (lines 744-788, 862-920)
- url_handler.py: Fixed variant detection (lines 633-665)
- discovery_service.py: Two-phase discovery (lines 137-214)
- 2 new comprehensive test files
Frontend:
- progress/types/progress.ts: Updated types with new fields
- progress/components/CrawlingProgress.tsx: Added UI sections
Real-world testing: Crawling supabase.com/docs now discovers
/docs/llms.txt and automatically follows 8 linked llms.txt files,
indexing complete documentation from all files.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Remove the special case that gave robots.txt sitemap declarations highest
priority, which incorrectly overrode the global priority order. Now properly
respects the intended priority: llms-full.txt > llms.txt > llms.md > llms.mdx >
sitemap.xml > robots.txt.
This fixes the issue where supabase.com/docs would return sitemap.xml instead
of llms.txt even though both files exist at /docs/ and llms.txt should have
higher priority.
Changes:
- Removed robots.txt early return that bypassed priority order
- Updated test to verify llms files take precedence over robots.txt sitemaps
- All discovery now follows consistent DISCOVERY_PRIORITY order
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Improve discovery logic to check the same directory as the base URL first before
falling back to root-level and subdirectories. This ensures files like
https://supabase.com/docs/llms.txt are found when crawling
https://supabase.com/docs.
Changes:
- Check same directory as base_url first (e.g., /docs/llms.txt for /docs URL)
- Fall back to root-level urljoin behavior
- Include base directory name in subdirectory checks (e.g., /docs subdirectory)
- Maintain priority order: same-dir > root > subdirectories
- Log discovery location for better debugging
This addresses cases where documentation directories contain their own llms.txt
or sitemap files that should take precedence over root-level files.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Preserve URL case in robots.txt parsing by only lowercasing the sitemap: prefix check
- Add support for relative sitemap paths in robots.txt using urljoin()
- Fix HTML meta tag parsing to use case-insensitive regex instead of lowercasing content
- Add URL scheme validation for discovered sitemaps (http/https only)
- Fix discovery target domain filtering to use discovered URL's domain instead of input URL
- Clean up whitespace and improve dict comprehension usage
These changes improve discovery reliability and prevent URL corruption while maintaining
backward compatibility with existing discovery behavior.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- add trailing slashes to agent-work-orders endpoints to prevent FastAPI mount() redirects
- add defensive null check for repository_url in detail view
- fix backend routes to use relative paths with app.mount()
- resolves ERR_NAME_NOT_RESOLVED when accessing agent work orders
Completes the implementation of test/review workflows with automatic resolution
and integrates them into the orchestrator.
**Phase 3: Test Workflow with Resolution**
- Created test_workflow.py with automatic test failure resolution
- Implements retry loop with max 4 attempts (configurable via MAX_TEST_RETRY_ATTEMPTS)
- Parses JSON test results and resolves failures one by one
- Uses existing test.md and resolve_failed_test.md commands
- Added run_tests() and resolve_test_failure() to workflow_operations.py
**Phase 4: Review Workflow with Resolution**
- Created review_workflow.py with automatic blocker issue resolution
- Implements retry loop with max 3 attempts (configurable via MAX_REVIEW_RETRY_ATTEMPTS)
- Categorizes issues by severity (blocker/tech_debt/skippable)
- Only blocks on blocker issues - tech_debt and skippable allowed to pass
- Created review_runner.md and resolve_failed_review.md commands
- Added run_review() and resolve_review_issue() to workflow_operations.py
- Supports screenshot capture for UI review (configurable via ENABLE_SCREENSHOT_CAPTURE)
**Phase 5: Compositional Integration**
- Updated workflow_orchestrator.py to integrate test and review phases
- Test phase runs between commit and PR creation (if ENABLE_TEST_PHASE=true)
- Review phase runs after tests (if ENABLE_REVIEW_PHASE=true)
- Both phases are optional and controlled by config flags
- Step history tracks test and review execution results
- Proper error handling and logging for all phases
**Supporting Changes**
- Updated agent_names.py to add REVIEWER constant
- Added configuration flags to config.py for test/review phases
- All new code follows structured logging patterns
- Maintains compatibility with existing workflow steps
**Files Changed**: 19 files, 3035+ lines
- New: test_workflow.py, review_workflow.py, review commands
- Modified: orchestrator, workflow_operations, agent_names, config
- Phases 1-2 files (worktree, state, port allocation) also staged
The implementation is complete and ready for testing. All phases now support
parallel execution via worktree isolation with deterministic port allocation.
- Enable SSL certificate verification (verify=True) for all HTTP requests
- Implement streaming with size limits (10MB default) to prevent memory exhaustion
- Add _read_response_with_limit() helper for secure response reading
- Update all test mocks to support streaming API with iter_content()
- Fix test assertions to expect new security parameters
- Enforce deterministic rounding in progress mapper tests
Security improvements:
- Prevents MITM attacks through SSL verification
- Guards against DoS via oversized responses
- Ensures proper resource cleanup with response.close()
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Resolved merge conflicts by integrating features from both branches:
- Added page_storage_ops service initialization from main
- Merged link text extraction with discovery mode features
- Preserved discovery single-file mode and domain filtering
- Maintained link text fallbacks for title extraction
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
* fix: Set explicit PLAYWRIGHT_BROWSERS_PATH to fix browser installation
Fixes Playwright browser not found error during web crawling.
The issue was introduced in the uv migration (9f22659) where the
browser installation path was not explicitly set as a persistent
environment variable.
Changes:
- Add ENV PLAYWRIGHT_BROWSERS_PATH=/ms-playwright
- Add --with-deps flag to playwright install command
- Add comprehensive root cause analysis document
Without this fix, Playwright installed browsers to a default location
at build time but couldn't find them at runtime, causing crawling
operations to fail with "Executable doesn't exist" errors.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
* fix: Remove --with-deps flag to prevent build conflicts
The --with-deps flag was causing build failures on some systems because:
- We already manually install all Playwright dependencies (lines 26-49)
- --with-deps attempts to reinstall these packages
- This causes package conflicts and build failures on Windows/WSL
The core fix (ENV PLAYWRIGHT_BROWSERS_PATH) remains the same.
* Delete PLAYWRIGHT_FIX_ANALYSIS.md
---------
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: Cole Medin <cole@dynamous.ai>
* Initial commit for RAG by document
* Phase 2
* Adding migrations
* Fixing page IDs for chunk metadata
* Fixing unit tests, adding tool to list pages for source
* Fixing page storage upsert issues
* Max file length for retrieval
* Fixing title issue
* Fixing tests
* fix: implement CASCADE DELETE for source deletion timeout issue
- Add migration 009 to add CASCADE DELETE constraints to foreign keys
- Simplify delete_source() to only delete parent record
- Database now handles cascading deletes efficiently
- Fixes timeout issues when deleting sources with thousands of pages
* chore: update complete_setup.sql to include CASCADE DELETE constraints
- Add ON DELETE CASCADE to foreign keys in initial setup
- Include migration 009 in the migrations tracking
- Ensures new installations have CASCADE DELETE from the start
Updates crawl4ai dependency to latest stable version with performance
and stability improvements.
Key improvements in 0.7.4:
- LLM-powered table extraction with intelligent chunking
- Fixed dispatcher bug for better concurrent processing
- Resolved browser manager race conditions
- Enhanced URL processing and proxy support
All existing tests pass (18/18). No breaking changes identified.
API remains backward compatible.
⚠️ IMPORTANT: URL Resolution Bug Status
A critical bug in v0.6.2 where ../../ paths only go up ONE directory
instead of TWO has been documented (see crawler-test branch). Status
in v0.7.4 is UNKNOWN - testing required before production deployment.
Test script provided: python/test_url_resolution_fix.py
Related issues fixed in v0.7.x:
- #570: General relative URL handling
- #1268: URLs after redirects
- #1323: Trailing slash base URL handling
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
* Add Anthropic and Grok provider support
* feat: Add crucial GPT-5 and reasoning model support for OpenRouter
- Add requires_max_completion_tokens() function for GPT-5, o1, o3, Grok-3 series
- Add prepare_chat_completion_params() for reasoning model compatibility
- Implement max_tokens → max_completion_tokens conversion for reasoning models
- Add temperature handling for reasoning models (must be 1.0 default)
- Enhanced provider validation and API key security in provider endpoints
- Streamlined retry logic (3→2 attempts) for faster issue detection
- Add failure tracking and circuit breaker analysis for debugging
- Support OpenRouter format detection (openai/gpt-5-nano, openai/o1-mini)
- Improved Grok provider empty response handling with structured fallbacks
- Enhanced contextual embedding with provider-aware model selection
Core provider functionality:
- OpenRouter, Grok, Anthropic provider support with full embedding integration
- Provider-specific model defaults and validation
- Secure API connectivity testing endpoints
- Provider context passing for code generation workflows
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* fully working model providers, addressing securtiy and code related concerns, throughly hardening our code
* added multiprovider support, embeddings model support, cleaned the pr, need to fix health check, asnyico tasks errors, and contextual embeddings error
* fixed contextual embeddings issue
* - Added inspect-aware shutdown handling so get_llm_client always closes the underlying AsyncOpenAI / httpx.AsyncClient while the loop is still alive, with defensive logging if shutdown happens late (python/src/server/services/llm_provider_service.py:14, python/src/server/ services/llm_provider_service.py:520).
* - Restructured get_llm_client so client creation and usage live in separate try/finally blocks; fallback clients now close without logging spurious Error creating LLM client when downstream code raises (python/src/server/services/llm_provider_service.py:335-556). - Close logic now sanitizes provider names consistently and awaits whichever aclose/close coroutine the SDK exposes, keeping the loop shut down cleanly (python/src/server/services/llm_provider_service.py:530-559). Robust JSON Parsing - Added _extract_json_payload to strip code fences / extra text returned by Ollama before json.loads runs, averting the markdown-induced decode errors you saw in logs (python/src/server/services/storage/code_storage_service.py:40-63). - Swapped the direct parse call for the sanitized payload and emit a debug preview when cleanup alters the content (python/src/server/ services/storage/code_storage_service.py:858-864).
* added provider connection support
* added provider api key not being configured warning
* Updated get_llm_client so missing OpenAI keys automatically fall back to Ollama (matching existing tests) and so unsupported providers still raise the legacy ValueError the suite expects. The fallback now reuses _get_optimal_ollama_instance and rethrows ValueError(OpenAI API key not found and Ollama fallback failed) when it cant connect. Adjusted test_code_extraction_source_id.py to accept the new optional argument on the mocked extractor (and confirm its None when present).
* Resolved a few needed code rabbit suggestion - Updated the knowledge API key validation to call create_embedding with the provider argument and removed the hard-coded OpenAI fallback (python/src/server/api_routes/knowledge_api.py). - Broadened embedding provider detection so prefixed OpenRouter/OpenAI model names route through the correct client (python/src/server/ services/embeddings/embedding_service.py, python/src/server/services/llm_provider_service.py). - Removed the duplicate helper definitions from llm_provider_service.py, eliminating the stray docstring that was causing the import-time syntax error.
* updated via code rabbit PR review, code rabbit in my IDE found no issues and no nitpicks with the updates! what was done: Credential service now persists the provider under the uppercase key LLM_PROVIDER, matching the read path (no new EMBEDDING_PROVIDER usage introduced). Embedding batch creation stops inserting blank strings, logging failures and skipping invalid items before they ever hit the provider (python/src/server/services/embeddings/embedding_service.py). Contextual embedding prompts use real newline characters everywhereboth when constructing the batch prompt and when parsing the models response (python/src/server/services/embeddings/contextual_embedding_service.py). Embedding provider routing already recognizes OpenRouter-prefixed OpenAI models via is_openai_embedding_model; no further change needed there. Embedding insertion now skips unsupported vector dimensions instead of forcing them into the 1536-column, and the backoff loop uses await asyncio.sleep so we no longer block the event loop (python/src/server/services/storage/code_storage_service.py). RAG settings props were extended to include LLM_INSTANCE_NAME and OLLAMA_EMBEDDING_INSTANCE_NAME, and the debug log no longer prints API-key prefixes (the rest of the TanStack refactor/EMBEDDING_PROVIDER support remains deferred).
* test fix
* enhanced Openrouters parsing logic to automatically detect reasoning models and parse regardless of json output or not. this commit creates a robust way for archons parsing to work throughly with openrouter automatically, regardless of the model youre using, to ensure proper functionality with out breaking any generation capabilities!
* updated ui llm interface, added seprate embeddings provider, made the system fully capabale of mix and matching llm providers (local and non local) for chat & embeddings. updated the ragsettings.tsx ui mainly, along with core functionality
* added warning labels and updated ollama health checks
* ready for review, fixed som error warnings and consildated ollama status health checks
* fixed FAILED test_async_embedding_service.py
* code rabbit fixes
* Separated the code-summary LLM provider from the embedding provider, so code example storage now forwards a dedicated embedding provider override end-to-end without hijacking the embedding pipeline. this fixes code rabbits (Preserve provider override in create_embeddings_batch) suggesting
* - Swapped API credential storage to booleans so decrypted keys never sit in React state (archon-ui-main/src/components/
settings/RAGSettings.tsx).
- Normalized Ollama instance URLs and gated the metrics effect on real state changes to avoid mis-counts and duplicate
fetches (RAGSettings.tsx).
- Tightened crawl progress scaling and indented-block parsing to handle min_length=None safely (python/src/server/
services/crawling/code_extraction_service.py:160, python/src/server/services/crawling/code_extraction_service.py:911).
- Added provider-agnostic embedding rate-limit retries so Google and friends back off gracefully (python/src/server/
services/embeddings/embedding_service.py:427).
- Made the orchestration registry async + thread-safe and updated every caller to await it (python/src/server/services/
crawling/crawling_service.py:34, python/src/server/api_routes/knowledge_api.py:1291).
* Update RAGSettings.tsx - header for 'LLM Settings' is now 'LLM Provider Settings'
* (RAG Settings)
- Ollama Health Checks & Metrics
- Added a 10-second timeout to the health fetch so it doesn't hang.
- Adjusted logic so metric refreshes run for embedding-only Ollama setups too.
- Initial page load now checks Ollama if either chat or embedding provider uses it.
- Metrics and alerts now respect which provider (chat/embedding) is currently selected.
- Provider Sync & Alerts
- Fixed a sync bug so the very first provider change updates settings as expected.
- Alerts now track the active provider (chat vs embedding) rather than only the LLM provider.
- Warnings about missing credentials now skip whichever provider is currently selected.
- Modals & Types
- Normalize URLs before handing them to selection modals to keep consistent data.
- Strengthened helper function types (getDisplayedChatModel, getModelPlaceholder, etc.).
(Crawling Service)
- Made the orchestration registry lock lazy-initialized to avoid issues in Python 3.12 and wrapped registry commands
(register, unregister) in async calls. This keeps things thread-safe even during concurrent crawling and cancellation.
* - migration/complete_setup.sql:101 seeds Google/OpenRouter/Anthropic/Grok API key rows so fresh databases expose every
provider by default.
- migration/0.1.0/009_add_provider_placeholders.sql:1 backfills the same rows for existing Supabase instances and
records the migration.
- archon-ui-main/src/components/settings/RAGSettings.tsx:121 introduces a shared credentialprovider map,
reloadApiCredentials runs through all five providers, and the status poller includes the new keys.
- archon-ui-main/src/components/settings/RAGSettings.tsx:353 subscribes to the archon:credentials-updated browser event
so adding/removing a key immediately refetches credential status and pings the corresponding connectivity test.
- archon-ui-main/src/components/settings/RAGSettings.tsx:926 now treats missing Anthropic/OpenRouter/Grok keys as
missing, preventing stale connected badges when a key is removed.
* - archon-ui-main/src/components/settings/RAGSettings.tsx:90 adds a simple display-name map and reuses one red alert
style.
- archon-ui-main/src/components/settings/RAGSettings.tsx:1016 now shows exactly one red banner when the active provider
- Removed the old duplicate Missing API Key Configuration block, so the panel no longer stacks two warnings.
* Update credentialsService.ts default model
* updated the google embedding adapter for multi dimensional rag querying
* thought this micro fix in the google embedding pushed with the embedding update the other day, it didnt. pushing now
---------
Co-authored-by: Chillbruhhh <joshchesser97@gmail.com>
Co-authored-by: Claude <noreply@anthropic.com>
* Preparing migration folder for the migration alert implementation
* Migrations and version APIs initial
* Touching up update instructions in README and UI
* Unit tests for migrations and version APIs
* Splitting up the Ollama migration scripts
* Removing temporary PRPs
---------
Co-authored-by: Rasmus Widing <rasmus.widing@gmail.com>
* Add Anthropic and Grok provider support
* feat: Add crucial GPT-5 and reasoning model support for OpenRouter
- Add requires_max_completion_tokens() function for GPT-5, o1, o3, Grok-3 series
- Add prepare_chat_completion_params() for reasoning model compatibility
- Implement max_tokens → max_completion_tokens conversion for reasoning models
- Add temperature handling for reasoning models (must be 1.0 default)
- Enhanced provider validation and API key security in provider endpoints
- Streamlined retry logic (3→2 attempts) for faster issue detection
- Add failure tracking and circuit breaker analysis for debugging
- Support OpenRouter format detection (openai/gpt-5-nano, openai/o1-mini)
- Improved Grok provider empty response handling with structured fallbacks
- Enhanced contextual embedding with provider-aware model selection
Core provider functionality:
- OpenRouter, Grok, Anthropic provider support with full embedding integration
- Provider-specific model defaults and validation
- Secure API connectivity testing endpoints
- Provider context passing for code generation workflows
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* fully working model providers, addressing securtiy and code related concerns, throughly hardening our code
* added multiprovider support, embeddings model support, cleaned the pr, need to fix health check, asnyico tasks errors, and contextual embeddings error
* fixed contextual embeddings issue
* - Added inspect-aware shutdown handling so get_llm_client always closes the underlying AsyncOpenAI / httpx.AsyncClient while the loop is still alive, with defensive logging if shutdown happens late (python/src/server/services/llm_provider_service.py:14, python/src/server/ services/llm_provider_service.py:520).
* - Restructured get_llm_client so client creation and usage live in separate try/finally blocks; fallback clients now close without logging spurious Error creating LLM client when downstream code raises (python/src/server/services/llm_provider_service.py:335-556). - Close logic now sanitizes provider names consistently and awaits whichever aclose/close coroutine the SDK exposes, keeping the loop shut down cleanly (python/src/server/services/llm_provider_service.py:530-559). Robust JSON Parsing - Added _extract_json_payload to strip code fences / extra text returned by Ollama before json.loads runs, averting the markdown-induced decode errors you saw in logs (python/src/server/services/storage/code_storage_service.py:40-63). - Swapped the direct parse call for the sanitized payload and emit a debug preview when cleanup alters the content (python/src/server/ services/storage/code_storage_service.py:858-864).
* added provider connection support
* added provider api key not being configured warning
* Updated get_llm_client so missing OpenAI keys automatically fall back to Ollama (matching existing tests) and so unsupported providers still raise the legacy ValueError the suite expects. The fallback now reuses _get_optimal_ollama_instance and rethrows ValueError(OpenAI API key not found and Ollama fallback failed) when it cant connect. Adjusted test_code_extraction_source_id.py to accept the new optional argument on the mocked extractor (and confirm its None when present).
* Resolved a few needed code rabbit suggestion - Updated the knowledge API key validation to call create_embedding with the provider argument and removed the hard-coded OpenAI fallback (python/src/server/api_routes/knowledge_api.py). - Broadened embedding provider detection so prefixed OpenRouter/OpenAI model names route through the correct client (python/src/server/ services/embeddings/embedding_service.py, python/src/server/services/llm_provider_service.py). - Removed the duplicate helper definitions from llm_provider_service.py, eliminating the stray docstring that was causing the import-time syntax error.
* updated via code rabbit PR review, code rabbit in my IDE found no issues and no nitpicks with the updates! what was done: Credential service now persists the provider under the uppercase key LLM_PROVIDER, matching the read path (no new EMBEDDING_PROVIDER usage introduced). Embedding batch creation stops inserting blank strings, logging failures and skipping invalid items before they ever hit the provider (python/src/server/services/embeddings/embedding_service.py). Contextual embedding prompts use real newline characters everywhereboth when constructing the batch prompt and when parsing the models response (python/src/server/services/embeddings/contextual_embedding_service.py). Embedding provider routing already recognizes OpenRouter-prefixed OpenAI models via is_openai_embedding_model; no further change needed there. Embedding insertion now skips unsupported vector dimensions instead of forcing them into the 1536-column, and the backoff loop uses await asyncio.sleep so we no longer block the event loop (python/src/server/services/storage/code_storage_service.py). RAG settings props were extended to include LLM_INSTANCE_NAME and OLLAMA_EMBEDDING_INSTANCE_NAME, and the debug log no longer prints API-key prefixes (the rest of the TanStack refactor/EMBEDDING_PROVIDER support remains deferred).
* test fix
* enhanced Openrouters parsing logic to automatically detect reasoning models and parse regardless of json output or not. this commit creates a robust way for archons parsing to work throughly with openrouter automatically, regardless of the model youre using, to ensure proper functionality with out breaking any generation capabilities!
---------
Co-authored-by: Chillbruhhh <joshchesser97@gmail.com>
Co-authored-by: Claude <noreply@anthropic.com>
- Changed default Ollama URL from localhost:11434 to host.docker.internal:11434
- This allows Docker containers to connect to Ollama running on the host machine
- Updated in backend services, frontend components, migration scripts, and documentation
- Most users run Archon in Docker but Ollama as a local binary, making this a better default
