MCP Containerd Server

What does “MCP Containerd” MCP Server do?

The MCP Containerd server is an implementation of the Model Context Protocol (MCP) designed to interface directly with Containerd’s CRI (Container Runtime Interface) using the Rust RMCP library. It allows AI assistants and clients to manage container workloads programmatically, enabling actions such as creating, starting, stopping, and deleting containers or pods, as well as interacting with container images. By exposing Containerd’s runtime and image services via standardized MCP endpoints, MCP Containerd empowers AI-driven workflows to automate container lifecycle management, perform image operations, and query status—all while integrating seamlessly with LLMs and AI agents. This enhances development and operations workflows by making complex container management accessible through structured, automated, and AI-assisted interactions.

List of Prompts

No prompt templates are mentioned in the repository.

List of Resources

No explicit MCP resources are described in the repository.

List of Tools

• version service: Provides CRI (Container Runtime Interface) version information.
• runtime service: Exposes container and pod runtime operations, including creating, starting, stopping, and deleting pods/containers, querying status, and executing commands in containers.
• image service: Offers image operations such as listing images, retrieving image status, pulling, and deleting images, and getting image filesystem information.

Use Cases of this MCP Server

• Container lifecycle management: Automate the creation, starting, stopping, and deletion of containers and pods within a Containerd environment, streamlining DevOps and testing workflows.
• Image management: Pull, list, and delete container images programmatically, enabling CI/CD pipelines or AI-driven environment preparation.
• Querying container status: Retrieve real-time status and diagnostics for pods and containers, useful for monitoring, troubleshooting, and intelligent orchestration.
• Executing commands in containers: Run arbitrary commands inside running containers, supporting remote debugging or automated maintenance tasks.
• Pod management: Manage pod sandboxes, including creation, status checks, and deletion, which benefits Kubernetes and microservices deployment scenarios.

How to set it up

Windsurf

1. Ensure you have Rust, Containerd, and Protobuf compilation tools installed.
2. Build the server:
   cargo build --release
3. Run the server:
   cargo run --release
4. Add MCP Containerd to your Windsurf configuration file.
5. Use this JSON snippet in your configuration:

"mcpServers": {
  "containerd-mcp": {
    "command": "cargo",
    "args": ["run", "--release"]
  }
}

Claude

1. Install Rust and Containerd.
2. Build the MCP Containerd server:
   cargo build --release
3. Configure Claude to connect to the running MCP server.
4. Add the following to your Claude configuration:

"mcpServers": {
  "containerd-mcp": {
    "command": "cargo",
    "args": ["run", "--release"]
  }
}

Cursor

1. Ensure prerequisites: Rust, Containerd.
2. Build with:
   cargo build --release
3. Launch the server:
   cargo run --release
4. Insert this snippet into Cursor’s configuration:

"mcpServers": {
  "containerd-mcp": {
    "command": "cargo",
    "args": ["run", "--release"]
  }
}

Cline

1. Install Rust, Containerd, and ensure Protobuf tools are available.
2. Build and run the server as above.
3. Add to your Cline config:

"mcpServers": {
  "containerd-mcp": {
    "command": "cargo",
    "args": ["run", "--release"]
  }
}

Securing API Keys

If your setup requires secrets (e.g., for future authentication), use environment variables:

"mcpServers": {
  "containerd-mcp": {
    "command": "cargo",
    "args": ["run", "--release"],
    "env": {
      "CONTAINERD_API_KEY": "${CONTAINERD_API_KEY}"
    },
    "inputs": {
      "api_key": "${CONTAINERD_API_KEY}"
    }
  }
}

How to use this MCP inside flows

Using MCP in FlowHunt

To integrate MCP servers into your FlowHunt workflow, start by adding the MCP component to your flow and connecting it to your AI agent:

Click on the MCP component to open the configuration panel. In the system MCP configuration section, insert your MCP server details using this JSON format:

{
  "containerd-mcp": {
    "transport": "streamable_http",
    "url": "https://yourmcpserver.example/pathtothemcp/url"
  }
}

Once configured, the AI agent is now able to use this MCP as a tool with access to all its functions and capabilities. Remember to change “containerd-mcp” to whatever the actual name of your MCP server is and replace the URL with your own MCP server URL.

Overview

Short evaluation:
MCP Containerd offers a clear bridge between Containerd and MCP, with robust tool coverage for container/image management. However, the absence of prompt templates and explicit resources reduces its out-of-the-box flexibility. It’s well-suited for DevOps automation and AI-driven workflows, but documentation and resource support could be improved.

MCP Score

Overall rating: 6/10. The MCP Containerd server provides strong core functionality for container management via MCP, but lacks prompt templates, explicit resource definitions, and comprehensive configuration documentation that would make it easier to adopt and extend.