mempalace-mcp
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., "@mempalace-mcpsave a memory: my anniversary is June 5th"
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.
mempalace-mcp
An HTTP Model Context Protocol (MCP) gateway that wraps the mempalace Python CLI, providing per-user memory palaces over a stateful HTTP API.
What it does
Exposes an MCP endpoint at
POST /mcp/:userIdOn first request for a user, spins up a dedicated
mempalacesubprocess (a Python MCP server) and keeps it alive in memoryProxies all MCP
tools/listandtools/callrequests to that user's subprocess, isolating each user's data under/data/<userId>/.mempalaceStrips extra parameters injected by MCP clients (e.g. n8n's
chatMessage) before forwarding tool calls
Related MCP server: mempalace-mcp-http
Endpoints
Method | Path | Description |
|
| MCP endpoint (initialize or resume session) |
|
| MCP SSE stream for active session |
|
| Close an MCP session |
|
| Pre-warm a user's palace and return its status |
|
| Mine a chat transcript into the palace, then clean up the temp file |
|
| Server health + active palace/session counts |
Running
Docker Compose (recommended)
Create a docker-compose.yml:
version: "3.9"
services:
mempalace-mcp:
image: ghcr.io/thehaseebahmed/mempalace-mcp:latest
container_name: mempalace-mcp
user: "1000:1000"
ports:
- "3000:3000"
volumes:
- ./palace-data:/data
environment:
- BASE_DIR=/data
- TRANSFORMERS_CACHE=/data/.cache/transformers
- HF_HOME=/data/.cache/huggingface
restart: unless-stoppedThen run:
docker compose up -dData is persisted to a palace-data/ folder next to your docker-compose.yml (mapped to /data in the container).
Local dev
npm run devConfiguration
Env var | Default | Description |
|
| Port to listen on |
|
| Root directory for all user palace data |
|
| Python binary with |
Requirements
Node.js >= 20
Python 3.11 +
mempalacepip package (handled automatically in Docker)
This server cannot be deployed
Maintenance
Related MCP Connectors
Personal knowledge graph as an AI memory layer over MCP - read, save, and link your memories.
Shared, governed long-term memory for AI agents across tools and sessions via MCP and REST.
- mcpOAuthai.butlerbrain
Persistent memory for AI assistants. Save once; recall from Claude, ChatGPT, or any MCP client.
MCP server for mandates, delegation, policy-gated execution, credential grants, and audit.
Related MCP Servers
- AlicenseNot gradedqualityCmaintenanceCentralized agentic memory and MCP gateway for LLMs, enabling persistent memory recall, validation, and integration with upstream MCP servers.1Apache 2.0
- FlicenseNot gradedqualityDmaintenanceEnables remote access to the MemPalace MCP server via HTTP, supporting bearer token authentication and concurrent clients while exposing all mempalace tools.-
- AlicenseBqualityBmaintenanceMCP server that connects Codex and other MCP clients to a running Glyph Hold instance via its HTTP API, enabling memory and secret management.7MIT
- FlicenseNot gradedqualityCmaintenanceExposes mempalace memory tools over HTTP MCP for agent consumption.-