Provider REST API

12 actions

Connect Claude to Hetzner

Manage cloud servers, networks, volumes, and firewalls. 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: API token
Actions exposed: 12
Cost per call (typical): 1 credit
Adapter: Maintained by Toolspoke

Connected in three steps

1
Install Hetzner
Open the marketplace in your workspace, add Hetzner to the project your agents work in, and it appears on the gateway immediately.
2
Connect the credential
Authenticate with api token. 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. Hetzner 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 Hetzner asks for

API token. 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.

API tokenRequired: Cloud Console → your project → Security → API tokens → Generate API token. Read & Write is required for create_server and server_power.; 64-character project token

What Claude can do in Hetzner

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: 10Reads
Writes: 2Writes
Destructive: 0Destructive

Reads

Fetches data and changes nothing.

list_servers
List the servers in the project with status, type, datacenter, and public IPs. Call this first to get the numeric server id the other server actions need.
get_server
Fetch one server by id, including its attached volumes, private networks, and applied firewalls.
server_metrics
Read time-series metrics for one server - cpu (percent), disk (IOPS and bandwidth), or network (packets and bytes per second). Use it to diagnose load or saturation before resizing.
get_action
Poll a long-running action by id to see whether it is still running, succeeded, or errored. create_server and server_power both return an action id.
list_server_types
List the server types available to the project with their cores, memory, disk, CPU type, and architecture. Use it to pick the server_type value for create_server.
list_images
List bootable images. Defaults to Hetzner's system images (the OS list); pass type=snapshot or type=backup to see your own. Use the name for create_server.
list_locations
List the datacenter locations servers can be created in, with their city, country, and network zone.
list_ssh_keys
List the SSH keys stored in the project. Pass the names to create_server so the new machine is reachable.
list_networks
List the project's private networks with their IP ranges, subnets, routes, and attached server ids.
list_volumes
List block-storage volumes with their size, location, format, and the server each one is attached to.

Writes

Creates or updates something on the other side.

create_server
Provision a new cloud server. Requires a name, a server type (see list_server_types) and an image (see list_images); pass ssh_keys so the machine is reachable, otherwise Hetzner returns a one-time root password in the response. Returns the server plus the action to poll with get_action.
server_power
Change a server's power state: poweron boots it, shutdown asks the guest OS to halt via ACPI, poweroff cuts power immediately, reboot sends an ACPI restart, and reset is a hard restart. The server and its disks are preserved in every case. Returns an action to poll with get_action.

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 Hetzner 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 Hetzner, but nothing in it deletes or permanently alters anything.
It reaches no further than your credential: Toolspoke holds no access to Hetzner 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 Hetzner: 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 Hetzner. An agent has to ask.
It does not smooth over provider limits: Toolspoke does not retry, queue or back off around Hetzner's own rate limits. A call that Hetzner refuses comes back to the agent as a failed call.

Before you connect it

What can Claude do in Hetzner?

12 named actions: 10 that only read and 2 that write. They include list_servers, get_server and create_server. 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 Hetzner connector need?

API token. The connector asks for api token. 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 Hetzner 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 Hetzner connector not do?

The 12 actions above are the whole of it. A call to any other name is refused before it reaches Hetzner 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 Hetzner, but nothing in it deletes or permanently alters anything. Toolspoke holds no access to Hetzner 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 Hetzner. An agent has to ask. Toolspoke does not retry, queue or back off around Hetzner's own rate limits. A call that Hetzner 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 Hetzner'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 Hetzner?

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 infrastructure connectors

Browse all of them