GraphMemory MCP Server

Most monitoring and observability work still happens through a UI a human drives. GraphMemory MCP server moves it into the conversation instead. AI-assisted development MCP providing long-term, on-device "AI memory" for IDEs. Powered by Kuzu GraphDB and exposed via MCP server.

The short version

An AI-assisted, long-term memory system for IDEs, powered by Kuzu graph database. GraphMemory-IDE is an MCP (Model Context Protocol) server that provides semantic vector search, graph-based knowledge storage, and real-time analytics. It integrates with VSCode, Cursor, and Windsurf through dedicated IDE plugins.

• Graph-based memory storage — — Kuzu native graph database with semantic vector search (HNSW indexes, sentence-transformers embeddings)
• Codon-accelerated graph algorithms — — Optional native compilation via Codon for 10-100x speedups on centrality, community detection, path analysis, and similarity computations, with automatic Python/NetworkX fallback
• FastAPI backend — — Async API with JWT authentication (EdDSA/Ed25519), rate limiting, and security middleware
• Real-time analytics — — WebSocket and SSE streaming for live telemetry dashboards
• Streamlit dashboard — — Interactive visualization of graph metrics, user activity, and system health
• Multi-IDE plugin support — — Extensions for VSCode, Cursor, and Windsurf

Getting it running

Setup follows the usual MCP pattern — install or clone the server, register it in your client's configuration file, restart the client.

What it needs from you

Configuration is passed through the environment: JWT_SECRET_KEY, DATABASE_URL, REDIS_URL, KUZU_DB_PATH, CODON_LIB_PATH. Treat anything key-shaped as a real credential — scope it to the minimum the server needs, and rotate it if it ever lands in a shared config.

How it compares

Among the monitoring and observability options, the useful question is rarely "what can it do" but "what does it cost you to run" — permissions, credentials, and how much of your context its toolset consumes. It is maintained by elementalcollision; worth a glance at recent repository activity before you build anything load-bearing on it.

This entry was verified against GraphMemory's own documentation before publication; SyncDev keeps the directory reviewed rather than auto-generated.

Things to watch

• It runs with your machine's permissions. That is convenient and also the reason to think about what you point it at before you approve a tool call.
• Missing credentials fail quietly in some clients — if no tools show up, check the environment block first.
• Keep per-call confirmation enabled while you learn its behaviour; it is the cheapest safeguard you have.