davidgut1982/lore-mcp
🔌 MCP Serverdavidgut1982
A persistent knowledge layer for AI agents featuring multi-agent attribution and flexible database support.
Lore-MCP addresses the critical challenge of state persistence in AI agent systems. By implementing the Model Context Protocol, it provides a standardized interface for agents to read, write, and query structured information. The system is designed to act as a 'second brain' for agents, supporting distinct data structures such as knowledge bases for factual storage, investigation threads for logical reasoning chains, and journals for chronological logging. A standout feature is its multi-agent attribution capability, which tracks which specific agent contributed or modified information, facilitating better collaboration and debugging in multi-agent environments. The architecture is highly flexible, supporting local development via SQLite or production-grade scaling through PostgreSQL and Supabase. This modularity makes it an essential component for developers building sophisticated, long-running AI assistants that require reliable, cross-session memory and context management.
💡Highlights
- ├─Multi-agent attribution tracking
- ├─Supports SQLite, Postgres, Supabase
- └─Persistent KB and thread storage
🎯For
- ├─AI Agent Developers
- └─Backend Engineers