MCP Servers

Explore Docker's container images for MCP servers

MCP Servers

MCP servers are containerized applications that implement the Model Context Protocol, enabling AI assistants like Claude to access external tools and services. Docker has partnered with Anthropic to build and maintain container images for MCP servers, available on Docker Hub under the mcp/ namespace.

Available MCP Servers

Below you’ll find a list of MCP servers available on Docker Hub:

Time Server

Working

Provides date and time capabilities

Docker Hub

Fetch Server

Working

Enables web browsing and content retrieval

Docker Hub

Filesystem Server

Working

Provides filesystem access within mounted directories

Docker Hub

PostgreSQL Server

Working

Connects to PostgreSQL databases

Docker Hub

Git Server

Working

Provides Git operations

Docker Hub

SQLite Server

Working

Enables SQLite database operations

Docker Hub

GitHub Server

Working

Interacts with GitHub repositories and resources

Docker Hub

Benefits of Containerized MCP Servers

Using Docker containers for MCP servers offers several advantages:

  1. Isolation and Security: Each MCP server runs in its own container, providing proper isolation and security.
  2. Easy Deployment: Docker makes it simple to deploy MCP servers in any environment that supports containers.
  3. Configuration Flexibility: Configure MCP servers through environment variables, bind mounts, and other Docker features.
  4. Resource Management: Control the resources allocated to each MCP server.
  5. Scalability: Scale MCP servers up or down based on demand.

Getting Started with MCP Servers

To use MCP servers with Docker, you can follow these steps:

  1. Pull the desired MCP server image from Docker Hub: 
    docker pull mcp/time
  2. Create a docker-compose.yml file to configure multiple MCP servers: 
    services:
  time:
    image: mcp/time
  fetch:
    image: mcp/fetch
  fs:
    image: mcp/filesystem
    command:
      - /rootfs
    volumes:
      - .:/rootfs
  3. Run the containers: 
    docker-compose up -d
  4. Connect your AI assistant to the MCP servers according to your application’s requirements.

Using MCP Servers with Claude

When integrating MCP servers with Claude, you’ll need to:

  1. Configure the appropriate environment variables for authentication
  2. Specify the tools you want Claude to have access to
  3. Ensure proper connection between Claude and the MCP servers

For detailed instructions, visit the Getting Started section.

Contributing

If you’ve developed a new MCP server that you’d like to share with the community, we welcome your contributions!

Check out the Contributing Guidelines for information on how to submit your MCP server to this collection.