Claude/claude md mhyyoe7gu6p8y6fq 01 t3 co rftbgo rx nt zqa gmp8v #157
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Add detailed documentation to help AI assistants understand and work with the Memori codebase, including: - Complete architecture overview and component relationships - Detailed codebase structure with 73+ Python modules - Development workflows and code quality standards - Database schema design and multi-tenant patterns - LLM integration patterns (OpenAI, Anthropic, LiteLLM) - Security best practices and common pitfalls - Testing guidelines and debugging tips - Pull request checklist and contribution guide This guide serves as a comprehensive reference for AI assistants to understand conventions, make safe changes, and maintain code quality.
Add comprehensive MCP server to expose Memori's persistent memory capabilities to any MCP-compatible AI assistant like Claude Desktop. Features: - 6 MCP tools for memory operations (record, search, retrieve, stats, etc.) - 2 MCP resources for read-only data access - 2 MCP prompt templates for common workflows - Multi-tenant isolation with user_id/session_id support - Support for all Memori database backends (SQLite, PostgreSQL, MySQL) - Automatic memory processing with OpenAI integration Files added: - mcp/memori_mcp_server.py - Main MCP server implementation - mcp/README.md - Comprehensive documentation - mcp/QUICKSTART.md - 5-minute setup guide - mcp/claude_desktop_config.json - Configuration template - mcp/__init__.py - Package initialization - examples/mcp/basic_usage_example.py - Basic usage examples - examples/mcp/advanced_workflow_example.py - Advanced workflows Tools provided: 1. record_conversation - Store conversations with automatic processing 2. search_memories - Intelligent memory search with filtering 3. get_recent_memories - Retrieve recent context 4. get_memory_statistics - Memory analytics and insights 5. get_conversation_history - Access conversation history 6. clear_session_memories - Reset session context Configuration: - Added mcp optional dependency group to pyproject.toml - Requires: mcp>=1.0.0, fastmcp>=2.0.0 - Uses uv for dependency management - Environment-based configuration (MEMORI_DATABASE_URL, OPENAI_API_KEY) Usage: 1. Install uv package manager 2. Configure Claude Desktop with provided JSON 3. Restart Claude Desktop 4. Use hammer icon to access Memori tools This enables Claude and other MCP clients to have persistent, queryable memory across all conversations with full SQL database backing.
Add comprehensive support for multiple LLM providers in the Memori MCP server, with OpenRouter as the recommended option for accessing 100+ models. Features: - OpenRouter support (100+ models including Claude, GPT-4, Llama, Mistral) - Azure OpenAI support (enterprise deployments) - Custom OpenAI-compatible endpoints (Ollama, LM Studio, etc.) - Automatic provider detection from environment variables - Priority-based configuration (OpenRouter > Azure > Custom > OpenAI) Provider detection priority: 1. OpenRouter (OPENROUTER_API_KEY) - access to 100+ models 2. Azure OpenAI (AZURE_OPENAI_API_KEY) - enterprise 3. Custom endpoint (LLM_BASE_URL) - local/self-hosted 4. OpenAI (OPENAI_API_KEY) - default Files modified: - mcp/memori_mcp_server.py - Added _detect_llm_provider() function - mcp/README.md - Comprehensive LLM provider configuration guide - mcp/QUICKSTART.md - Quick provider selection guide - mcp/claude_desktop_config_openrouter.json - OpenRouter config template - examples/mcp/openrouter_example.py - Full OpenRouter usage example Environment variables added: - OPENROUTER_API_KEY - OpenRouter API key - OPENROUTER_MODEL - Model to use (default: openai/gpt-4o) - OPENROUTER_BASE_URL - API base URL (default: https://openrouter.ai/api/v1) - OPENROUTER_APP_NAME - Optional app name for rankings - OPENROUTER_SITE_URL - Optional site URL for rankings - AZURE_OPENAI_* - Azure OpenAI configuration - LLM_BASE_URL - Custom endpoint URL - LLM_API_KEY - Custom API key - LLM_MODEL - Custom model name Benefits: - Access 100+ models through OpenRouter (Claude, GPT-4, Llama, etc.) - Use free models (Llama 3.1 70B, Mistral, etc.) - Run local models with Ollama/LM Studio - Enterprise Azure OpenAI support - Cost optimization through model selection - No vendor lock-in Popular OpenRouter models: - anthropic/claude-3.5-sonnet - Best for structured tasks - openai/gpt-4o - OpenAI's fastest GPT-4 - meta-llama/llama-3.1-70b-instruct - FREE, open-source - google/gemini-pro-1.5 - Google Gemini - mistralai/mixtral-8x7b-instruct - Cost-effective This enables users to choose the best model for their needs and budget, from premium models to free open-source options.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.