Back to directory

mcp-memory-service

A self-hosted memory backend that gives AI agent pipelines such as LangGraph, CrewAI, AutoGen and Claude a shared, persistent memory with a causal knowledge graph, reachable over a REST API or MCP.

mcp-memory-service

mcp-memory-service is a self-hosted memory backend for AI agent pipelines. It exposes one memory store over multiple transports, including a framework-agnostic REST API, MCP, OAuth and a CLI with a dashboard, so agents built with LangGraph, CrewAI, AutoGen or connected through Claude Desktop can store and retrieve context in a few milliseconds instead of starting each run from zero.

Agents share a knowledge graph with typed edges between memories, such as causes, fixes and contradicts relationships, rather than only a flat list of facts. An agent-ID header tags memories by the agent that created them for scoped retrieval, a conversation ID can bypass deduplication for incremental storage, and server-sent events notify other agents in real time when memories are stored or deleted. Autonomous consolidation compresses older memories so long-running context does not simply fill up.

Embeddings run locally through ONNX, so memory stays on the user's own infrastructure.

More in Agent memory

See all 450
  • A long technical guide to AI agent memory: why stateless LLMs need it, the different memory architectures from vector stores to knowledge graphs, and how Cognee implements one as an open-source memory layer.

  • A local, file-based long-term memory layer for a personal AI that never summarizes conversations, keeping every line with a timestamp and searching by time first and words second, rather than retrieving only similar-seeming chunks.

  • NEOTH

    deepwiki.com

    A local-first personal AI daemon in Rust with five memory tiers, consent-gated tools, a WASM plugin sandbox, multi-provider model routing and signed audit logs, aimed at people who want both a daily assistant and a serious coding helper.