Provider REST API

12 actions

Connect Claude to Docker Engine

Inspect and restart containers, images, and Compose services on a Docker host. Toolspoke puts 12 of its actions behind one MCP endpoint that Claude, Cursor, and Codex all speak.

Connect View pricing

Connection: Provider REST API
Authentication: Engine API URL
Actions exposed: 12
Cost per call (typical): 1 credit
Adapter: Maintained by Toolspoke

Connected in three steps

1
Install Docker Engine
Open the marketplace in your workspace, add Docker Engine to the project your agents work in, and it appears on the gateway immediately.
2
Connect the credential
Authenticate with engine api url. Where to get one, and what it has to be able to reach, is the next section.
3
Point your agent at the gateway
Give your client one address, https://toolspoke.com/mcp. Claude Code takes it as a command, Claude and Claude Desktop add it as a custom connector, and Cursor, Codex and VS Code each read it from a config file of their own.

.mcp.json

{
  "mcpServers": {
    "toolspoke": {
      "type": "http",
      "url": "https://toolspoke.com/mcp"
    }
  }
}

One block covers every tool you have installed. Docker Engine shows up in the client as soon as your policy allows it, and so does everything else you install later.

Where the address goes, per client

Claude Code: Run it in your project, then /mcp to sign in
claude mcp add --transport http toolspoke https://toolspoke.com/mcp
Claude and Claude Desktop: Settings, then Connectors, then Add custom connector
https://toolspoke.com/mcp
Cursor: ~/.cursor/mcp.json, or .cursor/mcp.json for one project
{ "mcpServers": { "toolspoke": { "url": "https://toolspoke.com/mcp" } } }
Codex: ~/.codex/config.toml
[mcp_servers.toolspoke] url = "https://toolspoke.com/mcp"
VS Code: .vscode/mcp.json, or the MCP: Add Server command
{ "servers": { "toolspoke": { "type": "http", "url": "https://toolspoke.com/mcp" } } }

What Docker Engine asks for

Engine API URL. You provide it once, when you install the connector. Toolspoke encrypts it at rest and decrypts it only for the length of a single call, and the gateway attaches it to the outbound request itself, so it is never part of the arguments an agent sends.

Engine API URLRequired: A publicly reachable Docker Engine API endpoint, usually a socket proxy. Never expose an unauthenticated daemon.; https://docker.example.com:2376
Bearer tokenOptional: Sent as Authorization: Bearer when the endpoint sits behind an auth proxy
API versionOptional: Engine API version prefix. Use "none" to send unversioned paths.; 1.44

What Claude can do in Docker Engine

12 actions, each one declared and named by the connector rather than discovered at runtime. A workspace policy grants a person all of them, a hand-picked selection, everything on the read side, everything on the write side, or none.

Reads: 9Reads
Writes: 3Writes
Destructive: 0Destructive

Reads

Fetches data and changes nothing.

list_containers
List containers on the daemon with their image, state, uptime and published ports. Defaults to running containers only; set all to true to include stopped ones. Start here to find the container name that the other actions need.
inspect_container
Return the configuration and current state of one container: exit code, health check result, restart policy, resource limits, port bindings, mounts and network addresses. Use it after list_containers to explain why a container is unhealthy or restarting. Environment variables are deliberately omitted because they usually hold secrets.
container_logs
Read the tail of a container's logs, with stdout and stderr returned separately. This is the fastest way to diagnose a crashing or noisy container. Output is capped at roughly 20000 characters per stream, keeping the most recent lines.
container_stats
Sample one container's live CPU, memory, network and block I/O usage and return a computed summary, the same numbers `docker stats` shows. Takes a couple of seconds because the daemon needs two samples to derive CPU percentage.
container_processes
List the processes running inside a container, as `docker top` does. Use it to confirm whether the expected worker or server process is actually alive inside a container that looks healthy from the outside.
list_images
List images held on the daemon with their tags, size and creation date. Use it to check whether a deploy tag has actually been pulled, or to find images taking up disk.
inspect_image
Return the details of one image: entrypoint, command, exposed ports, labels, layer count, architecture and the digest it was pulled at. Use it to verify what a running tag actually contains.
system_info
Report the daemon's own health: engine version, OS and kernel, storage driver, CPU and memory available, and how many containers are running, paused or stopped. Call it when you need to know what the host itself looks like.
disk_usage
Summarise how much disk images, containers, local volumes and the build cache are using, as `docker system df` does. Use it before asking a human to reclaim space; this adapter never prunes anything itself.

Writes

Creates or updates something on the other side.

start_container
Start a stopped container. Returns a note instead of an error when the container is already running.
stop_container
Stop a running container with SIGTERM, escalating to SIGKILL after the grace period. The container keeps its filesystem and can be started again - this never removes it. Returns a note instead of an error when it is already stopped.
restart_container
Restart a container in place, the usual first remediation for a wedged or leaking service. The container and its data are preserved.

What it will not do

Enforced by the gateway rather than left to convention, which is why each of these can be stated flatly.

It cannot call anything else: The 12 actions above are the whole of it. A call to any other name is refused before it reaches Docker Engine rather than forwarded on, and connecting your account does not add to the list: it is fixed by the connector, not discovered at run time.
Nothing here deletes: This connector writes to Docker Engine, but nothing in it deletes or permanently alters anything.
It reaches no further than your credential: Toolspoke holds no access to Docker Engine of its own. Every call carries the credential you stored and nothing besides, so whatever that credential cannot reach, this connector cannot reach either.
It never hears from Docker Engine: Nothing is pushed to it. There is no webhook, no subscription and no polling, so this connector cannot notice by itself that something changed in Docker Engine. An agent has to ask.
It does not smooth over provider limits: Toolspoke does not retry, queue or back off around Docker Engine's own rate limits. A call that Docker Engine refuses comes back to the agent as a failed call.

Before you connect it

What can Claude do in Docker Engine?

12 named actions: 9 that only read and 3 that write. They include list_containers, inspect_container and container_logs. Nothing outside that list is reachable: the connector declares each operation by name rather than proxying whatever an agent asks for.

What credentials does the Docker Engine connector need?

Engine API URL. The connector asks for engine api url, and optionally bearer token and api version. Values are encrypted at rest and attached to the outbound request by the gateway, so they are never part of the arguments an agent sends and never reach the audit log.

Does the Docker Engine connector work with Cursor and Codex, or only Claude?

Any client that speaks MCP, and every one of them gets the same 12 actions. There is a single address, https://toolspoke.com/mcp. Claude Code adds it with claude mcp add --transport http, Claude and Claude Desktop take it as a custom connector in settings, Cursor reads it from .cursor/mcp.json, Codex from ~/.codex/config.toml, and VS Code from .vscode/mcp.json. Each of them signs in to the gateway itself, so there is no key to paste.

What does the Docker Engine connector not do?

The 12 actions above are the whole of it. A call to any other name is refused before it reaches Docker Engine rather than forwarded on, and connecting your account does not add to the list: it is fixed by the connector, not discovered at run time. This connector writes to Docker Engine, but nothing in it deletes or permanently alters anything. Toolspoke holds no access to Docker Engine of its own. Every call carries the credential you stored and nothing besides, so whatever that credential cannot reach, this connector cannot reach either. Nothing is pushed to it. There is no webhook, no subscription and no polling, so this connector cannot notice by itself that something changed in Docker Engine. An agent has to ask. Toolspoke does not retry, queue or back off around Docker Engine's own rate limits. A call that Docker Engine refuses comes back to the agent as a failed call.

Can I limit which actions an agent can call?

Yes, in two places. The project switches Docker Engine's actions on and off one at a time, for everyone in the project at once, and the screen groups them by read, write and destructive so turning off everything that deletes is one click. An individual agent key can then be narrowed further, to particular toolkits in a project and to particular actions in a toolkit. Whatever it was granted, a key never reaches a project its owner cannot.

What gets recorded when an agent calls Docker Engine?

Every attempt, with the agent that made it and the person that agent belongs to, the full request payload, the response payload, the status, the duration, and the credits spent. Values whose key names a secret are masked out before the record is shown to anyone. An operation the connector marks as not retained never has its response body written at all, so the gateway keeps no second copy of what was read.

More developer tools connectors

Browse all of them