Skip to main content
Glama
claude-proxmox-controller

claude-proxmox-controller

console_record

Record a running guest's console as an MP4/H.264 video at its own resolution, including VM sound when available, without showing it on your screen; start, stop, or check recording status.

Instructions

Record a running guest's console as a video (MP4, H.264), with the VM's sound when it has a SPICE sound card, at the guest's own resolution, with nothing shown on the user's screen. action start begins it — then work in the console as usual (console_send, console_click…): typing, commands and builds all go into the video — and action stop finishes it and says where the file is (the user's Videos folder, Proxmox Controller). status says whether one runs. It stops by itself after max_minutes (60 unless given) or when the guest stops. Works on desktops, servers' text screens and containers (containers without sound).

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": "console_record"
  }
}

Input Schema

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

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

A4.5/5.0
Behavior4/5

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

The description goes well beyond the annotations: it discloses output format (MP4/H.264), that recording is invisible on the user's screen, that sound requires a SPICE sound card, the auto-stop default of 60 minutes, the file's destination, and container/headless-screen support. Gaps remain for a mutation tool — it doesn't say what happens if start is called while already recording, or if stop is called with nothing running.

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

Conciseness4/5

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

It is a single dense paragraph that front-loads the core purpose before the action semantics and edge cases. All information earns its place, though the em-dash run-on structure makes it somewhat harder to scan than a short bulleted breakdown would.

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?

For a mutation tool with no output schema, it covers output location, lifecycle, defaults, platform coverage and sound caveats — enough to call it correctly. It omits error/failure behavior (start-when-recording, stop-when-idle), which is the main remaining hole given no output schema exists to describe returns.

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

Parameters4/5

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

Schema coverage is 67%, and the description adds real value on the undocumented parts: max_minutes gets a default ('60 unless given') that the schema omits, sound is qualified by the SPICE sound-card condition, and each action value is defined behaviorally. host, vmid and reason are still only covered by the schema.

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?

The description names a specific verb and resource (record a guest's console as MP4/H.264 video) and pins down scope: guest resolution, sound only with a SPICE sound card, nothing shown on the user's screen. It also disambiguates the three action values so an agent can tell this apart from console_screenshot or console_send immediately.

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

Usage Guidelines5/5

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

It explicitly sequences usage: 'action start begins it — then work in the console as usual (console_send, console_click…)' and 'action stop finishes it', with status as the check. It also states the implicit termination conditions (max_minutes, guest stop), which tells the agent when the tool stops being needed without any inference.

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