mcps
D

Recipe

DevOps managing containers with Docker MCP

Status, logs, restart, scale — through chat. No more `docker ps | grep | awk` reflexes.

for Devs and DevOps who keep forgetting docker syntax and want a status check + restart in one message

Hand-checked recipe · curated by Ilya Gindin

10 minutes

to run

intermediate

difficulty

1

connector

What you'll need

Docker Connects to your Docker daemon to read state and execute container commands.
Required

This recipe runs in 4 steps

  1. 1

    Connect Docker MCP

    Say to your AI

    Install Docker MCP — it connects to your local Docker daemon.

    You'll get: Claude confirms it can list running containers and images.

  2. 2

    Check container status

    Say to your AI

    Show me all running containers and their memory usage.

    You'll get: A small table with container name, image, status, and current memory.

  3. 3

    Debug issues

    Say to your AI

    The api-service container keeps crashing — show me the last 50 log lines and diagnose the issue.

    You'll get: Recent log excerpt + a one-paragraph diagnosis with the most likely cause and a suggested fix.

  4. 4

    Manage services

    Say to your AI

    Restart the redis container and verify it comes back up healthy.

    You'll get: Container restart confirmation + a health check (port responding, status running).

You're done when

Container management through natural language — perfect for developers who forget docker syntax between Tuesday and Thursday.

Why this workflow exists

Local Docker setups drift. Something is running, something else isn’t, the logs you need are five containers deep. Docker MCP turns “what’s wrong with my stack” from a series of muscle-memory commands into one chat turn.