docker_mcp_server

Provides an MCP server that exposes Docker operations as Model Context Protocol (MCP) tools via an HTTP/SSE/WebSocket bridge. It connects to a Docker daemon (optionally over TLS) and uses MCP Mediator to generate MCP tools for Docker client service methods, including many container/image/swarm/network/volume/registry operations.

Evaluated Apr 04, 2026 (16d ago)
Repo ↗ DevTools mcp docker devtools automation containers mcp-server
⚙ Agent Friendliness
51
/ 100
Can an agent use this?
🔒 Security
35
/ 100
Is it safe for agents?
⚡ Reliability
19
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
78
Documentation
70
Error Messages
0
Auth Simplicity
55
Rate Limits
5

🔒 Security

TLS Enforcement
60
Auth Strength
25
Scope Granularity
10
Dep. Hygiene
45
Secret Handling
40

Security posture is heavily dependent on how the service is deployed. The MCP bridge endpoints appear to be accessible over HTTP/WebSocket/SSE without documented authentication/authorization; tool coverage includes highly privileged Docker operations. TLS options relate to connecting to the Docker daemon (e.g., --tls-verify, --cert-path), not necessarily to securing the MCP endpoints. Mounting /var/run/docker.sock grants broad host-level control. Registry auth is supported for push/pull, but no guidance is provided for secret storage/log redaction.

⚡ Reliability

Uptime/SLA
0
Version Stability
30
Breaking Changes
20
Error Recovery
25
AF Security Reliability

Best When

Used locally or inside a tightly controlled environment (dev machine, dedicated build agent, internal network) with limited exposure, where operators can trust the agent and its prompts.

Avoid When

Avoid exposing endpoints to the public internet or any broad audience; avoid using with weak/no TLS between components where traffic could be intercepted; avoid if you cannot sandbox/limit the Docker daemon permissions available to the service.

Use Cases

  • Giving an AI agent controlled, tool-based access to a local/remote Docker daemon (e.g., create/start/stop containers).
  • CI/CD automation where an agent needs to inspect images/containers and manage swarm resources.
  • Interactive operations assistance for developers (e.g., diagnose container state, view logs, inspect networks/volumes).
  • Registry tasks via Docker (pull/push images; authenticate).

Not For

  • Publicly exposed deployments without strong network/auth controls (the tool set can take destructive actions).
  • Use cases requiring fine-grained RBAC or per-tool authorization guarantees (not described).
  • Environments where mounting the Docker socket or granting Docker daemon access is prohibited.
  • Multi-tenant agent platforms where users could trigger host-impacting Docker operations.

Interface

REST API
Yes
GraphQL
No
gRPC
No
MCP Server
Yes
SDK
No
Webhooks
No

Authentication

Methods: No first-class auth described for the MCP bridge endpoints (only Docker daemon connectivity options). Docker registry authentication is supported as a Docker command/tool (docker_authenticate), which implies registry credentials are needed for that tool.
OAuth: No Scopes: No

README focuses on Docker daemon connection (DOCKER_HOST, TLS verify, cert path) rather than authenticating requests to the MCP/HTTP/WebSocket/SSE endpoints. No MCP endpoint authentication/authorization mechanism or scope model is documented.

Pricing

Free tier: No
Requires CC: No

No pricing info; this appears to be a self-hosted open-source component (GPL-3.0).

Agent Metadata

Pagination
none
Idempotent
False
Retry Guidance
Not documented

Known Gotchas

  • The MCP tools include destructive operations (remove containers/images/services, prune, leave swarm, etc.); an agent needs strong guardrails/confirmation logic.
  • If the Docker daemon is reachable via mounted docker.sock or exposed TCP, the MCP server effectively becomes a powerful remote-control surface.
  • Non-annotated method tool generation is supported, and tool naming/descriptions may be inferred—agents should still validate tool semantics and expected parameters.
  • Tool behavior depends on the permissions of the Docker daemon identity used by the server; failures may surface as Docker API errors without documented MCP-specific remediation.

Alternatives

Full Evaluation Report

Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for docker_mcp_server.

AI-powered analysis · PDF + markdown · Delivered within 30 minutes

$99

Package Brief

Quick verdict, integration guide, cost projections, gotchas with workarounds, and alternatives comparison.

Delivered within 10 minutes

$3

Score Monitoring

Get alerted when this package's AF, security, or reliability scores change significantly. Stay ahead of regressions.

Continuous monitoring

$3/mo

Scores are editorial opinions as of 2026-04-04.

8642
Packages Evaluated
17761
Need Evaluation
586
Need Re-evaluation
Community Powered