OpenMemory
Local persistent memory store for LLM applications including claude desktop, github copilot, codex, antigravity, etc.
OpenMemory is an MCP server that provides persistent, local-first long-term memory capabilities for AI agents and LLM applications. Unlike traditional RAG solutions that rely solely on vector similarity, it implements a multi-sector cognitive memory engine with episodic, semantic, procedural, emotional, and reflective memory types. The server runs self-hosted with SQLite or Postgres, offers Python and Node SDKs, and integrates with frameworks like LangChain, CrewAI, and AutoGen while exposing MCP tools for Claude, Cursor, and Windsurf.
Key Features
Use Cases
- 01Adding persistent memory to Claude Desktop conversations so context survives across sessions
- 02Building AI coding assistants in Cursor or Windsurf that remember project patterns and user preferences
- 03Creating multi-agent systems with CrewAI or AutoGen that share a long-term knowledge base
- 04Developing personal AI journaling or note-taking applications with evolving user context
- 05Implementing customer support chatbots that retain interaction history and preferences per user
- 06Constructing research assistants that track temporal facts and how knowledge evolves over time
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.
OpenMemory — FAQ
What is OpenMemory MCP server?+
OpenMemory is an MCP server that provides long-term cognitive memory for AI agents, supporting multi-sector memory types (episodic, semantic, procedural, emotional, reflective) with temporal reasoning. It runs locally with SQLite by default and exposes memory operations as MCP tools for Claude, Cursor, Windsurf, and other MCP clients.
How do I install OpenMemory MCP server?+
Clone the repository, configure the .env file, and run the backend server with npm or Docker. For MCP clients like Claude Desktop or Cursor, add the HTTP transport pointing to http://localhost:8080/mcp in your client's MCP configuration file.
Which AI clients work with OpenMemory?+
OpenMemory works with any MCP-aware client including Claude Desktop, Claude Code, Cursor, and Windsurf. It also provides Python and Node SDKs for direct integration with LangChain, CrewAI, AutoGen, and Streamlit applications.
Does OpenMemory require API keys?+
OpenMemory can run completely offline with local SQLite and synthetic embeddings. For production use, you may optionally configure OpenAI, OpenRouter, Gemini, Ollama, or AWS embeddings by setting the appropriate API keys in the .env file.
Is OpenMemory free and open source?+
Yes, OpenMemory is licensed under Apache 2.0 and is completely free to use, self-hosted, and open source. You own the database and can run it entirely offline without vendor lock-in.
What are the prerequisites for running OpenMemory?+
You need Node.js (for the backend server) or Python 3.x (for the SDK). The server uses SQLite by default (no separate database setup required), but you can optionally configure Postgres for multi-user deployments.
How do I install OpenMemory?+
Open the source repository on GitHub and follow its README. OpenMemory is a mcp server — MCP Agents Market links you directly to the official repo.
Is OpenMemory free?+
OpenMemory is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.