Feat:Openrouter/Anthropic/grok-support

[Chillbruhhh]

Pull Request

Summary

Added OpenRouter LLM provider support and EMBEDDING_PROVIDER to database setting that allows different embeddings model to be selected

Changes Made

• Added OpenRouter as LLM provider option - Access to 200+ models from Anthropic, OpenAI, Meta, Google, etc.
• Added EMBEDDING_PROVIDER setting in database migration
• Enhanced credential service validation with auto-recovery for missing provider settings
• Updated comprehensive documentation across RAG guides, configuration docs, and getting started
• Added mixed provider support - Use OpenRouter for LLM + OpenAI/Google/Ollama for embeddings
• Improved test coverage with OpenRouter-specific embedding fallback tests

Type of Change

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Documentation update
• Performance improvement
• Code refactoring

Affected Services

• Frontend (React UI)
• Server (FastAPI backend)
• MCP Server (Model Context Protocol)
• Agents (PydanticAI service)
• Database (migrations/schema)
• Docker/Infrastructure
• Documentation site

Testing

• All existing tests pass
• Added new tests for new functionality
• Manually tested affected user flows
• Docker builds succeed for all services

Test Evidence

# Backend tests with new OpenRouter embedding fallback tests
cd python && python -m pytest tests/test_async_llm_provider_service.py -v

# Frontend tests
cd archon-ui-main && npm run test

# Integration test - RAG operations with OpenRouter LLM + OpenAI embeddings
# Tested crawling: https://docs.langchain.com/llms.txt
# Result: ✅ No "Unsupported LLM provider: false" errors

Checklist

• My code follows the service architecture patterns
• If using an AI coding assistant, I used the CLAUDE.md rules
• I have added tests that prove my fix/feature works
• All new and existing tests pass locally
• My changes generate no new warnings
• I have updated relevant documentation
• I have verified no regressions in existing features

Breaking Changes

Add this table to your database if you havent! (added to the migration table)

-- Added EMBEDDING_PROVIDER setting
INSERT INTO archon_settings (key, value, is_encrypted, category, description) VALUES
('EMBEDDING_PROVIDER', 'openai', false, 'rag_strategy', 'Embedding provider to use: openai, ollama, or google')
ON CONFLICT (key) DO NOTHING;

Additional Notes

Key Technical Details:

• OpenRouter Limitation: OpenRouter doesn't provide embedding models, so the system intelligently falls back to OpenAI's embedding API when OpenRouter is selected for embeddings
• Auto-Recovery: If EMBEDDING_PROVIDER setting is missing, the system automatically creates it with safe defaults
• Mixed Provider Architecture: Users can now use OpenRouter for chat models while using OpenAI, Google, or Ollama for embeddings
• Database Migration: Safe migration with ON CONFLICT (key) DO NOTHING for existing installations

Documentation Updated:

• RAG configuration guide with OpenRouter section
• Getting started guide with provider options
• Configuration documentation with mixed provider setup
• README feature list and prerequisites

Recommended Setup for OpenRouter Users:

• LLM Provider: OpenRouter (for access to 200+ models)
• Embedding Provider: OpenAI (for reliable embeddings)
• Requires both OpenRouter and OpenAI API keys

Summary by CodeRabbit

• New Features

  • Added OpenRouter and Grok provider support with provider-specific defaults.
  • Introduced server-side provider connectivity checks and status indicators in settings.
  • Persisted per-provider chat/embedding model selections across sessions.
  • Enabled provider-aware code extraction and summarization flows.

• Improvements

  • Smarter embedding routing based on selected model (Google/OpenAI).
  • Enhanced contextual embeddings with provider-aware defaults and parameter handling.
  • Updated provider UI: ordering, styling, alerts, and visibility.
  • More reliable health checks and fallback behavior across providers.
  • Faster, periodic credential and connectivity re-validation.