Docker MCP
Docker MCP is a Docker image that wraps the mcp-server-docker Python package with supergateway, exposing a Model Context Protocol (MCP) server over HTTP/SSE…
Platform availability
Available on 1 of 10 platforms
- Unraid
- TrueNAS (not listed)
- Umbrel (not listed)
- ZimaOS (not listed)
- Proxmox (not listed)
- Helm (not listed)
- Mac (not listed)
- Windows (not listed)
- Linux (not listed)
- Any server (not listed)
Health score?A 0-100 score computed nightly from four factors: maintained (40%: recent commits and steady activity), popular (25%: GitHub stars, log scale), easy to install (20%: platforms, install notes, setup guides, Docker image) and light to run (15%: minimum RAM, ARM support). When the source code has been checked, a fifth factor, security check (20%: automated scan for committed secrets, unsafe desktop-app settings, known-vulnerable dependencies and project hygiene; any serious finding caps it at 20), is added. Factors without data are left out and the rest are rescaled.
39/100 Low
- Maintained
- 69last commit 65 days ago
- Popular
- 00 GitHub stars
- Easy to install
- 351 install platform, install notes, Docker image
- Light to run
- 304 GB minimum RAM
Availability: platform listings confirmed today, links checked 24 days ago
- GitHub stars
- 0
- Open issues
- 0
- Last commit
- 2026-09-17
- Activity
- stale
Checked 8 days ago - source: GitHub (louij2/docker-mcp)
Resources & Compatibility
Minimum requirements: 4 GB RAM, 6 GB disk. ARM (e.g. Raspberry Pi) is not supported.
Checked 54 days ago - source: Docker Hub image tags and published docs
First-install notes
- Mark containers for Autoheal Add the autoheal label set to true on each service you want monitored, or set AUTOHEAL_CONTAINER_LABEL=all to watch every container. Without this label, Autoheal will ignore the container and will not restart it when it stops working. source
- Containers need health checks Give each monitored container a health check so Docker can report whether it is healthy or unhealthy. Autoheal only restarts containers it has detected as unhealthy, so a missing or broken check means no automatic recovery. source
- Run Autoheal with Docker access Start the Autoheal container with /var/run/docker.sock mounted and restart: always, because that gives it permission to see and restart other containers. Without that Docker socket access, Autoheal cannot actually fix anything. source
- Security The image mounts docker.sock, which is root-equivalent on the host, so expose the MCP endpoint only to trusted clients. source
- Build Build this base image first before building the other louij2 MCP images. source
Fetched 39 days ago - each note links to its source
Variants
Alternatives
No alternatives collected yet.
Common questions
How should I limit who can reach the container?
It mounts the Docker socket, which is root-equivalent on the host, so only run it where you trust every client that can reach it.
How does a stdio MCP server become available to HTTP clients?
The image wraps the mcp-server-docker Python package with supergateway, bridging the stdio MCP server over HTTP.
What failure areas should I check if the setup does not work?
Common failure modes to check include Docker environment issues, gateway configuration problems, tool registration failures, secret mismatches, and container crashes.
Which Docker operations can the server perform?
It uses standard Docker CLI commands to list, run, stop, and remove containers, as well as to pull images from registries.
Can I build other MCP server images on top of this one?
Yes, several sibling images build FROM local/docker-mcp:latest and layer their own MCP server on top. Build this image first if you are rebuilding that family from scratch.
Answers sourced from deepwiki.com, github.com, ubos.tech
Community
- MCP gateway and protocols - Docker Community Forums (forum)
- Latest MCP Catalog topics - Docker Community Forums (forum)
Collected 53 days ago - public community threads