Openmemory
БесплатноНе проверенA FastMCP server that exposes mem0 long-term memory to MCP clients over SSE, enabling AI agents to store and recall facts persistently.
Описание
A FastMCP server that exposes mem0 long-term memory to MCP clients over SSE, enabling AI agents to store and recall facts persistently.
README
A FastMCP server that exposes mem0 long-term memory to MCP clients (e.g. an AI coding agent) over SSE.
It is a thin, stateless MCP layer over an existing mem0 REST server
(http://mem0:8000) — it does not run its own vector store. This keeps a
single source of truth for memory (the mem0 pgvector store) instead of
introducing a second, disconnected memory silo.
Tools
| Tool | Purpose |
|---|---|
search_memory(query, user_id?, limit?) |
Recall memories relevant to a query. Call at the start of a task. |
add_memory(text, user_id?, role?) |
Store a fact/preference/exchange. Use automatically when you learn something durable, or on explicit request. |
list_memories(user_id?) |
List stored memories for a bucket. |
delete_memory(memory_id) |
Forget a single memory by id. |
Memory is isolated per user_id ("bucket"). When a caller omits user_id,
MEM0_DEFAULT_USER_ID is used, giving the agent a single shared memory.
Configuration
Environment variables (prefix MEM0_):
| Variable | Default | Description |
|---|---|---|
MEM0_MEM0_URL |
http://mem0:8000 |
Base URL of the mem0 REST server |
MEM0_API_KEY |
(none) | Admin API key, sent as X-API-Key |
MEM0_DEFAULT_USER_ID |
coder-agent |
Bucket used when no user_id is passed |
MEM0_SEARCH_LIMIT |
5 |
Default number of memories per search |
MEM0_TIMEOUT |
10 |
HTTP timeout (s) for reads (search/list) |
MEM0_WRITE_TIMEOUT |
60 |
HTTP timeout (s) for writes (add runs a server-side LLM extraction) |
MEM0_SERVER_HOST |
0.0.0.0 |
Bind host for the SSE server |
MEM0_SERVER_PORT |
8000 |
Bind port for the SSE server |
Run
pip install -e .
openmemory-mcp-sse # SSE at http://0.0.0.0:8000/sse
# or
openmemory-mcp # stdio mode
Container
Built by the release-containers workflow (release-please) to
ghcr.io/dedsxc/openmemory-mcp:<version> for linux/amd64,linux/arm64.
Установить Openmemory в Claude Desktop, Claude Code, Cursor
unyly install mcp-openmemoryСтавит в Claude Desktop, Claude Code, Cursor и VS Code — сам разбирается с npx, uvx и сборкой из исходников.
Впервые? Поставь CLI: curl -fsSL https://unyly.org/install | sh
Или настроить вручную
Выполни в терминале:
claude mcp add mcp-openmemory -- uvx openmemory-mcpFAQ
Openmemory MCP бесплатный?
Да, Openmemory MCP бесплатный — установка в пару кликов через Unyly без оплаты.
Нужен ли API-ключ для Openmemory?
Нет, Openmemory работает без API-ключей и переменных окружения.
Openmemory — hosted или self-hosted?
Доступен hosted-вариант: Unyly запускает сервер в облаке, локальная установка не обязательна.
Как установить Openmemory в Claude Desktop, Claude Code или Cursor?
Открой Openmemory на unyly.org, выбери вкладку своего клиента (Claude Desktop, Claude Code, Cursor) и нажми Install — конфиг сгенерируется автоматически, без правки JSON.
Похожие MCP
Fetch
Web content fetching and conversion for efficient LLM usage.
AWS KB Retrieval
Retrieval from AWS Knowledge Base using Bedrock Agent Runtime.
автор: modelcontextprotocolSpring AI MCP Server
Provides auto-configuration for setting up an MCP server in Spring Boot applications.
llm-analysis-assistant
A very streamlined mcp client that supports calling and monitoring stdio/sse/streamableHttp, and can also view request responses through the /logs page. It also
автор: xuzexin-hzCompare Openmemory with
Не уверен что выбрать?
Найди свой стек за 60 секунд
Автор?
Embed-бейдж для README
Похожее
Все в категории ai