mcp-memory-service
Open-source persistent memory for AI agent pipelines (LangGraph, CrewAI, AutoGen) and Claude. REST API + knowledge graph + autonomous consolidation.
MCP Memory Service is an open-source persistent memory backend for AI agents that operates as both an MCP server and REST API. It enables AI assistants to retain context across sessions using semantic search, knowledge graphs, and local embeddings that run in 5ms. The service works with 25+ AI tools including Claude Desktop, Claude Code, LangGraph, CrewAI, and AutoGen, storing memories locally or syncing via Cloudflare. It features autonomous memory consolidation, entity extraction, and a web dashboard for visualization and management.
Key Features
Use Cases
- 01Persistent context for Claude Desktop or Claude Code sessions that survives restarts
- 02Shared memory layer for multi-agent systems (LangGraph, CrewAI, AutoGen clusters)
- 03Team collaboration with cloud-synced memories across devices via Cloudflare backend
- 04Browser-based AI memory in claude.ai using Remote MCP over HTTPS
- 05Knowledge base for coding assistants to remember project architecture and past decisions
- 06Inter-agent messaging bus using tagged memories as coordination signals
Related MCP Servers
View moremarkitdown
Python tool for converting files and office documents to Markdown.
firecrawl
The context API to search, scrape, and interact with the web at scale. 🔥
prompts.chat
f.k.a. Awesome ChatGPT Prompts. Share, discover, and collect prompts from the community. Free and open source — self-host for your organization with complete privacy.
langflow
Langflow is a powerful tool for building and deploying AI-powered agents and workflows.
mcp-memory-service — FAQ
What is MCP Memory Service?+
It's an open-source persistent memory backend that works as both an MCP server and REST API, allowing AI agents and assistants to store and retrieve context across sessions using semantic search and knowledge graphs.
How do I install it for Claude Desktop?+
Install via pip (pip install mcp-memory-service), then add a 'memory' entry to your Claude Desktop config JSON with command 'memory' and args ['server']. Restart Claude Desktop to activate.
Which AI clients does it work with?+
It supports 25+ tools including Claude Desktop, Claude Code, Cursor, VS Code, LangGraph, CrewAI, AutoGen, OpenCode, Windsurf, and claude.ai (via Remote MCP). Any MCP-compatible client or HTTP client can use it.
Do I need API keys or cloud services?+
No. Embeddings run locally via ONNX with no external API calls. Cloud sync (Cloudflare) is optional; the default SQLite backend is fully offline and free.
Is it free and open source?+
Yes, it's Apache 2.0 licensed with zero runtime costs. All processing happens locally unless you opt into the optional Cloudflare sync backend.
Can multiple agents share the same memory store?+
Yes. Agents can share one local database or cloud backend, using X-Agent-ID headers and tags to scope or share memories across runs.
How do I install mcp-memory-service?+
Open the source repository on GitHub and follow its README. mcp-memory-service is a mcp server — MCP Agents Market links you directly to the official repo.
Is mcp-memory-service free?+
mcp-memory-service is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.