Provides a persistent semantic memory server for AI tools using pgvector and Supabase.
Sponsored
Brain Kit offers an open-source Model Context Protocol (MCP) memory server, empowering AI tools like Claude, Cursor, and ChatGPT with a persistent semantic memory. Built on Supabase with pgvector for efficient semantic search, it helps AI agents retain context across sessions, eliminating the need for users to repeatedly provide background information. This self-hostable engine allows developers to manage their AI's memory with powerful thought capture and retrieval capabilities, running frugally at minimal cost.
Key Features
01Browse and filter stored memories by type, topic, person, or days
02Automated capture of new thoughts with auto-generated embeddings and metadata extraction
03API endpoint for programmatic thought capture via webhooks
04Semantic search for past thoughts using pgvector's HNSW index
05Generate statistics on captured thoughts, topics, and people
060 GitHub stars
Use Cases
01Enable advanced semantic search of historical AI interactions, personal notes, and captured content.
02Equip AI assistants and chatbots with a long-term memory to maintain context across sessions.
03Integrate a cost-effective, self-hosted memory backend for custom AI applications and agents.
