Skip to main content
Glama
claude-proxmox-controller

claude-proxmox-controller

guest_power

Start, cleanly shut down, force stop, or reboot a Proxmox VM or container; use force stop only when shutdown fails.

Instructions

Start, shut down (clean), stop (at once, like pulling the plug — only if shut down does not work) or reboot a guest.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "host": "<host>",
      "vmid": "<vmid>",
      "action": "<action>",
      "reason": "<reason>"
    },
    "name": "guest_power"
  }
}

Input Schema

NameRequiredDescriptionDefault
hostYesThe host's id, from list_hosts.
vmidYesThe guest's ID (VMID).
actionYes
reasonYesWhy you are doing this, in a sentence. The user sees it in the approval window and in the log.

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

A4.2/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations only declare non-read-only, non-destructive, non-idempotent; the description adds the important behavioral distinction that shutdown is clean while stop is an immediate cut. The 'like pulling the plug' warning is genuine risk context beyond structured data, though it sits in mild tension with destructiveHint=false and nothing is said about what a stopped or rebooting guest does to in-flight work.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

One sentence, front-loaded with the four actions, with parentheticals that each carry a real distinction (clean vs immediate) and a conditional rule. Nothing is padding and nothing needed is cut.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

All four required parameters are documented and no output schema exists, so the description's job is mostly action semantics, which it covers. Gaps are minor: no note on prerequisites (guest must exist, host reachable) or what state the guest is left in after a failed stop.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 75% and host, vmid and reason are already self-documented there, including the approval-window note. The description's real contribution is explaining what each action value means, which the bare enum cannot; that is useful but it is restating the name-level verbs rather than adding format, preconditions, or per-action requirements.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States specific verbs (start, shut down, stop, reboot) against a specific resource (a guest), and even disambiguates the two superficially similar stop modes by contrast with a plug pull. An agent can tell this apart from host_power or bulk_power from the resource word alone, and the action enum maps one-to-one onto the described verbs.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives explicit intra-tool guidance: use the forced 'stop' only when 'shutdown' does not work, which establishes an escalation order. It does not say when to reach for this versus bulk_power (many guests) or host_power (whole host), so routing to siblings is left implicit rather than stated.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.