Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@RAG-mcp what did I tell you about the RAG setup yesterday?"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Memória persistente para chat: salva conversas e busca as mais relevantes por similaridade (SQLite + FAISS).
memory_store.py — salva cada turno em SQLite e indexa embeddings no FAISS (embeddings vêm de um servidor llama.cpp).
server_mcp.py — servidor MCP que expõe as ferramentas:
store_memory — salva um turno (input do usuário + output do assistente)
remember_memory — busca memórias semanticamente parecidas com o contexto
remember_memory_date — busca todas as memórias de um dia (YYYY-MM-DD)
Related MCP server: MCPMem
Python 3
Um servidor llama.cpp rodando com endpoint de embeddings (padrão: http://127.0.0.1:8001)
pip install -r requirements.txtpython server_mcp.pyVariáveis de ambiente opcionais:
Variável | Padrão |
|
|
|
|
Os dados ficam salvos em ./.llama_memhistory/ (banco SQLite + índice FAISS).
MIT
Memory system for AI agents with semantic search. Store and recall memories with ease.
Long-term memory for AI assistants. Hybrid retrieval, query expansion, auto-topics.
Long-term memory for AI assistants. Isolated per-user storage, recall across conversations.
Persistent memory for AI agents. Search and store durable facts, preferences and decisions.
