mcp-server-docker  by ckreiling

Docker manager with natural language interface

created 7 months ago
550 stars

Top 59.0% on sourcepulse

GitHubView on GitHub
Project Summary

This project provides a Docker server for the "MCP" (Meta-Command Protocol) framework, enabling users to manage Docker containers and resources using natural language prompts. It targets server administrators, tinkerers, and AI enthusiasts who want to leverage LLMs for infrastructure management. The primary benefit is simplifying complex Docker operations through intuitive conversational commands.

How It Works

The server acts as an intermediary between an LLM and the Docker daemon. It translates natural language requests into Docker API calls, allowing users to compose containers, manage volumes, inspect logs, and more, all via text prompts. This approach abstracts away the intricacies of Docker commands and configurations, making Docker management more accessible.

Quick Start & Requirements

  • Install:
    • Via PyPI with uv: uvx mcp-server-docker
    • Via Docker: Build image with docker build -t mcp-server-docker . and run with docker run -i --rm -v /var/run/docker.sock:/var/run/docker.sock mcp-server-docker:latest
  • Prerequisites: Docker installed and running. Mounting the Docker socket (/var/run/docker.sock) is required for the Docker-based installation to control the host's Docker daemon.
  • Demo: https://github.com/user-attachments/assets/65e35e67-bce0-4449-af7e-9f4dd773b4b3

Highlighted Details

  • Supports composing containers via natural language prompts, including a plan-and-apply loop.
  • Provides access to container stats (CPU, memory) and logs.
  • Offers tools for managing containers, images, networks, and volumes.
  • Remembers project state across chat sessions for easier cleanup and resumption.

Maintenance & Community

No specific contributors, sponsorships, or community links (Discord/Slack) are mentioned in the README.

Licensing & Compatibility

The repository does not explicitly state a license.

Limitations & Caveats

The project warns against configuring containers with sensitive data, as any data exchanged with the LLM is considered compromised unless the LLM runs locally. For security, it does not support privileged containers or capabilities manipulation (--cap-add/--cap-drop).

Health Check
Last commit

1 month ago

Responsiveness

1 day

Pull Requests (30d)
2
Issues (30d)
1
Star History
172 stars in the last 90 days

Explore Similar Projects

Starred by Elie Bursztein Elie Bursztein(Cybersecurity Lead at Google DeepMind), Tim J. Baek Tim J. Baek(Founder of Open WebUI), and
1 more.

harbor by av

0.3%
2k
CLI tool for local LLM stack orchestration
created 1 year ago
updated 3 weeks ago
Feedback? Help us improve.