mem0ai/mem0-mcp
Free管理代码偏好和模式，支持语义搜索，方便在 IDE 中存取技术文档。 (Mem0 官方)
FreeFree tier
Inputs: text
About mem0ai/mem0-mcp
mem0-mcp-server wraps the official Mem0 Memory API as a Model Context Protocol (MCP) server, enabling MCP-compatible clients (such as Claude Desktop, Cursor, custom agents) to add, search, update, and delete long-term memories. It exposes tools like add_memory, search_memories, get_memories, update_memory, delete_memory, delete_all_memories, delete_entities, and list_entities. Deployment options include Python package, Docker, or Smithery remote hosting. Note: This repository is archived and no longer actively maintained; a cloud-hosted MCP server is now officially available.
Key Features
add_memory: Save text, conversation history, or explicit message objects for a user/agent
search_memories: Semantic search across existing memories with filters and limit
get_memories: List memories with structured filters and pagination
get_memory: Retrieve a single memory by memory_id
update_memory: Overwrite a memory's text using its memory_id
delete_memory: Delete a single memory by memory_id
delete_all_memories: Bulk delete all memories in a confirmed scope (user/agent/app/run)
delete_entities: Delete a user/agent/app/run entity and its memories
list_entities: Enumerate users/agents/apps/runs stored in Mem0
Multiple deployment options: Python package, Docker container, or Smithery remote hosting
Pros & Cons
Pros
- Open source (MIT license) and freely available
- Works with any MCP-compatible client
- Simple setup via pip/uv, Docker, or Smithery
- Comprehensive set of memory management tools
- Semantic search capabilities with filters and pagination
Cons
- Repository archived and no longer actively maintained (as of Mar 2026)
- Requires a Mem0 API key (external dependency) to function
- Memory storage is limited to Mem0 API capabilities
- No offline/local memory storage option
Best For
Persistent memory for AI agents and chatbotsSemantic memory search across user/agent conversationsManaging and updating long-term memories in MCP-compatible clientsEnabling memory persistence across sessions in Claude Desktop, Cursor, or custom agentsBulk memory cleanup and entity management for applications
FAQ
What is mem0-mcp-server?
It is an MCP server that wraps the official Mem0 Memory API, allowing MCP-compatible clients to add, search, update, and delete long-term memories.
How do I install mem0-mcp-server?
Install via pip or uv: 'pip install mem0-mcp-server' or 'uv pip install mem0-mcp-server'. You can also use Docker or Smithery for deployment.
What tools does mem0-mcp-server expose?
It exposes add_memory, search_memories, get_memories, get_memory, update_memory, delete_memory, delete_all_memories, delete_entities, and list_entities.
Is mem0-mcp-server still actively maintained?
No, the repository was archived on March 24, 2026 and is now read-only. However, Mem0 offers an official cloud-hosted MCP server.
What are the deployment options?
You can run it as a Python package locally via uv/pip, as a Docker container with an HTTP endpoint, or use Smithery for a managed remote service.