MCP Memory Service

MCP Memory Service provides persistent memory for AI agent pipelines through an API and knowledge graph.

github.com/doobidoo/mcp-memory-service
MCP Memory Service homepage screenshot

About MCP Memory Service

MCP Memory Service is an open-source persistent memory backend for AI agent pipelines, including LangGraph, CrewAI, AutoGen, and Claude. It provides a REST API, knowledge graph with typed causal edges, and autonomous memory consolidation to let agents retain project context, architecture decisions, and code patterns across sessions. The service runs self-hosted with local ONNX embeddings so memory stays on user infrastructure, supports real-time SSE notifications for memory changes, and uses X-Agent-ID and conversation_id headers for scoped retrieval and incremental conversation storage.

Key features

  • Framework-agnostic REST API for agent memory operations
  • Knowledge graph with typed causal edges for shared context
  • Local ONNX embeddings to keep memory on user infrastructure
  • Autonomous consolidation to compress old memories
  • Real-time SSE notifications for memory store/delete events
  • Scoped retrieval via X-Agent-ID and conversation_id headers
Best for: Developers building AI agent pipelines with supported frameworks who need persistent, shared memory without cloud lock-in or external API costs.
Good for
  • Self-hosted deployment avoids cloud lock-in and extra API costs
  • Local ONNX embeddings keep memory data on user infrastructure
  • Shared knowledge graph enables causal context access across agent runs

MCP Memory Service at a glance

Type
Open-source project
Pricing
Not confirmed. Check the site
Open source
Yes · Source code
Language
English
Last checked
Oct 5, 2026 · Online when checked
Listed
Oct 5, 2026

MCP Memory Service alternatives

Tools and sites you could use instead, ranked by how closely they match.

All 24 alternatives

Popular in AI Agent Frameworks

Well-known listings from the same topic.

All 75

MCP Memory Service FAQ

Which AI agent frameworks are compatible with MCP Memory Service?
It supports LangGraph, CrewAI, AutoGen, Claude Desktop, Claude Code, OpenCode, and any HTTP client.
Where are memory embeddings processed?
Embeddings run locally via ONNX, so memory data stays on user infrastructure.
How can memory retrieval be scoped to specific agents or conversations?
Use the X-Agent-ID header to tag memories by agent, or conversation_id to bypass deduplication for incremental storage.
Is something wrong or out of date?

Tell us and an editor will check it.

Own this website?

Claim the listing and suggest updates yourself. An editor checks every claim first.

Claim this listing