About Mcp Memory Bank
Mcp Memory Bank is a context management MCP server for LLM applications. It uses ChromaDB and configurable embeddings to keep project-specific memory persistent, and can be deployed with Docker Compose.
Key features
- 01Persistent project-specific memory
- 02ChromaDB vector storage
- 03Configurable embedding model
- 04Docker Compose deployment
Use cases
- 01Retaining context for project conversations
- 02Running a local memory service with Docker
MCP support
It provides an MCP server for LLM applications, with persistent context backed by ChromaDB.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project README🐳 Running with Docker
This project is fully Docker-ready for easy deployment and local development. The provided Dockerfile and docker-compose.yml set up both the main application and its required ChromaDB vector database.
Requirements
- Docker (latest stable)
- Docker Compose (v2+ recommended)
Environment Variables
The following environment variables are used by default (can be overridden in your environment or via docker-compose.yml):
CHROMADB_URL=http://chromadb:8000
TRANSPORT=http
HTTP_PORT=3000
MCP_MEMBANK_EMBEDDING_MODEL=Xenova/all-MiniLM-L6-v2
NODE_ENV=production
NODE_OPTIONS=--max-old-space-size=4096
Build & Run
To build and start all services:
docker-compose up --build -d
This will:
- Build the main TypeScript application (Node.js v22.13.1-slim)
- Start the app as
ts-app(listening on port 3000) - Start ChromaDB as
chromadb(listening on port 8000) - Create a persistent volume for ChromaDB data
- Set up a shared Docker network for inter-service communication
Ports
- 3000: Main application HTTP API (
ts-app) - 8000: ChromaDB vector database (
chromadb)
Data Persistence
- ChromaDB data is persisted in the named Docker volume
chromadb-data. - Application data directory (
/app/data) is created and owned by a non-root user inside the container.
Special Notes
- The application requires ChromaDB to be available at the URL specified by
CHROMADB_URL(default:http://chromadb:8000). - The embedding model can be changed via the
MCP_MEMBANK_EMBEDDING_MODELenvironment variable. - If you need to customize environment variables, edit the
docker-compose.ymlor use an.envfile.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.