Docker Engine Cf MCP Server — Hosted Devops Integration
MCP Server language Hosted language PublicManage Docker containers, images, volumes, and networks across dev/stg/prod environments — via Cloudflare Tunnel relay.
Use with AI AssistantsMCP
Connect Claude, Cursor, or any MCP-compatible client — then call tools directly
① Add This MCP Server
Paste into your AI client config — then all its tools are available instantly.
{
"mcpServers": {
"docker-engine-cf": {
"url": "https://mcp.aerostack.dev/s/aerostack/mcp-docker-engine-cf",
"headers": {
"Authorization": "Bearer YOUR_AEROSTACK_TOKEN"
}
}
}
}Replace YOUR_AEROSTACK_TOKEN with your API token from the dashboard.
② Call a Tool
Ask your AI assistant to call a specific tool, or send raw JSON-RPC:
Natural Language Prompt
“Use the list_environments tool to list all configured docker environments (dev, stg, prod) and their relay urls”
Using a Workspace?
Add this MCP to your Workspace — your team shares one token, secrets are stored securely, and every AI agent in the workspace can call it without per-user setup.
description Overview
mcp-docker-engine-cf — Docker Engine MCP (Cloudflare Worker)
Manage containers, images, volumes, and networks across dev/stg/prod environments — from any Aerostack agent via Cloudflare Tunnel.
A Cloudflare Worker MCP that calls docker-relay servers running on your infrastructure behind Cloudflare Tunnel. No Docker ports need to be exposed publicly — all traffic flows through signed HTTPS requests over the tunnel.
Live endpoint: https://mcp.aerostack.dev/s/aerostack/mcp-docker-engine-cf
What You Can Do
- List and inspect containers with real-time CPU/memory stats
- Tail logs from any container with time and line filters
- Start, stop, restart, and remove containers
- Exec commands inside running containers for diagnostics
- Manage Docker images: list, pull, remove
- Inspect networks and volumes
- View Docker system info and disk usage
- Manage Docker Compose services: list, logs, restart
Architecture
Aerostack Agent
│
▼
mcp-docker-engine-cf (this Worker — on Cloudflare edge)
│ HTTPS + Bearer token
▼
Cloudflare Tunnel endpoint (e.g. docker-dev.yourdomain.com)
│
▼
docker-relay (127.0.0.1:4242 on your server)
│
▼
Docker Engine (local socket)
See docker-relay/README.md for per-server setup instructions.
Configuration
| Variable | Required | Description |
|---|---|---|
DOCKER_RELAY_SECRET |
Yes | Shared Bearer token — set same value in each docker-relay process and here |
DOCKER_DEV_URL |
No | Relay HTTPS URL for dev environment (e.g. https://docker-dev.yourdomain.com) |
DOCKER_STG_URL |
No | Relay HTTPS URL for staging environment |
DOCKER_PROD_URL |
No | Relay HTTPS URL for production environment |
At least one of DOCKER_DEV_URL, DOCKER_STG_URL, or DOCKER_PROD_URL must be set.
Quick Start
Add to Aerostack Workspace
- Go to aerostack.dev → Your Project → MCPs
- Search for "Docker Engine" and click Add to Workspace
- Add your credentials under Project → Secrets
Example Prompts
"List all running containers on prod"
"Show me the last 200 log lines from the api container on stg"
"What's CPU/memory usage across all containers on dev?"
"Restart the nginx service in the web compose project on prod"
"What images are on dev and when were they pulled?"
Direct API Call
curl -X POST https://mcp.aerostack.dev/s/aerostack/mcp-docker-engine-cf \
-H 'Content-Type: application/json' \
-H 'X-Mcp-Secret-DOCKER-RELAY-SECRET: your-relay-secret' \
-H 'X-Mcp-Secret-DOCKER-DEV-URL: https://docker-dev.yourdomain.com' \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"list_containers","arguments":{"environment":"dev"}}}'
Available Tools
| Tool | Description |
|---|---|
list_environments |
List configured environments and their relay URLs |
list_containers |
List containers with status, image, ports (all or filtered) |
get_container_logs |
Tail logs with line count, since filter, and timestamps |
inspect_container |
Full container details: config, env vars, mounts, health |
get_container_stats |
Live CPU, memory, and network stats |
start_container |
Start a stopped container |
stop_container |
Stop a running container |
restart_container |
Restart a container |
remove_container |
Remove a container (force option for running containers) |
exec_in_container |
Run a command inside a running container |
list_images |
List Docker images with size and age |
pull_image |
Pull an image from a registry |
remove_image |
Remove a local image |
list_networks |
List Docker networks |
inspect_network |
Get network details including connected containers |
list_volumes |
List Docker volumes |
inspect_volume |
Get volume details and mount point |
system_info |
Docker daemon info: version, resources, runtime |
disk_usage |
Docker disk usage breakdown (images, containers, volumes) |
list_compose_services |
List services in a Docker Compose project |
compose_logs |
Get logs from a Compose service |
compose_restart |
Restart one or all Compose services |
License
MIT
terminal Tools (22)
Available tools on this MCP server. Each tool can be called directly from any AI agent.
list_environments #1 List all configured Docker environments (dev, stg, prod) and their relay URLs
list_containers #2 List all containers with status, image, ports, and names
get_container_logs #3 Get logs from a container with optional tail and time filtering
inspect_container #4 Full container details: config, env vars, mounts, network, health status
get_container_stats #5 Real-time CPU, memory, network, and block I/O stats snapshot
start_container #6 Start a stopped container
stop_container #7 Stop a running container gracefully
restart_container #8 Restart a container
remove_container #9 Remove a stopped container
exec_in_container #10 Run a command inside a running container and return output
list_images #11 List Docker images with repository, tag, size, and creation date
pull_image #12 Pull a Docker image from a registry
remove_image #13 Remove a local Docker image
list_networks #14 List all Docker networks with driver, scope, and subnet info
inspect_network #15 Get Docker network details including connected containers
list_volumes #16 List all Docker volumes with driver and mount point
inspect_volume #17 Get details about a Docker volume including its mount path
system_info #18 Docker daemon info: version, OS, kernel, CPU/memory, container and image counts
disk_usage #19 Docker disk usage breakdown: images, containers, volumes, build cache
list_compose_services #20 List Docker Compose services and their status
compose_logs #21 Get logs from a Docker Compose service
compose_restart #22 Restart one or all services in a Docker Compose project
Details
language Live Endpoint
https://mcp.aerostack.dev/s/aerostack/mcp-docker-engine-cf
Sub-50ms globally · Zero cold start
Publisher
Pre-built functions for the most common MCP tool patterns. Clone, extend, and deploy.
More in Devops
Browse Devops MCPs →Asana Project Management
by @aerostack
Tasks, projects, sections, comments, teams, custom fields via Asana's official MCP
Atlassian Cloud
by @aerostack
Jira issues, sprints, Confluence pages via Atlassian's official MCP
Cloudflare Platform
by @aerostack
Cloudflare Workers, KV, R2, D1, Pages, DNS via Cloudflare's official MCP
Datadog Observability
by @aerostack
Dashboards, monitors, logs, metrics, incidents, SLOs, APM traces via Datadog's official MCP
Figma Design
by @aerostack
Files, frames, components, comments, variables via Figma's official MCP
GitHub API
by @aerostack
GitHub repos, PRs, issues, branches, code search via GitHub's official hosted MCP
Frequently asked questions
What is the Docker Engine Cf MCP server and what can it do? +
The Docker Engine Cf MCP server is hosted on Aerostack and exposes these tools to your AI agent: `list_environments`, `list_containers`, `get_container_logs`, `inspect_container`, `get_container_stats`. You get one hosted URL — no self-hosting — that works from Claude, Cursor, ChatGPT, Gemini, VS Code, or any MCP-compatible client, and you can share it with your team or combine it with other MCP servers in a workspace.
Is the Docker Engine Cf MCP server hosted, or do I have to run it myself? +
It's hosted on Aerostack's edge infrastructure — you don't deploy or maintain anything. Add it to a workspace and you get one authenticated URL, with secrets encrypted, that any AI agent or editor can connect to. Use it solo or share the same URL across your whole team.
Which AI agents and editors can use the Docker Engine Cf MCP server? +
Any MCP client: Claude and Claude Code, Cursor, ChatGPT, Gemini, Windsurf, Cline, VS Code, and custom agents. Because it's one hosted URL, the same Docker Engine Cf MCP server works everywhere — and you can compose it with other MCP servers, skills, and functions behind a single workspace URL.
How do I install the Docker Engine Cf MCP server in Claude Desktop? +
Add the following to your Claude Desktop config (`claude_desktop_config.json`): ```json { "mcpServers": { "@aerostack/mcp-docker-engine-cf": { "command": "npx", "args": ["-y", "@aerostack/@aerostack/mcp-docker-engine-cf"] } } } ``` Then restart Claude Desktop and the tools will appear automatically.
How do I use the Docker Engine Cf MCP server in Cursor? +
In Cursor, open **Settings → MCP** and add: ```json { "name": "@aerostack/mcp-docker-engine-cf", "command": "npx", "args": ["-y", "@aerostack/@aerostack/mcp-docker-engine-cf"] } ``` Save and reload Cursor. The MCP tools will be available in Agent mode.
Does Docker Engine Cf MCP require authentication? +
Yes. Docker Engine Cf requires authentication. Check the MCP's documentation for the required credentials.