agentleFS
Sign inStart free
trac3r00trac3r00/agent-skills/skills/proxmox-ve/SKILL.md

proxmox-ve

Skill0 starsChanged 3 months agoMITView on GitHub

Safely operate Proxmox VE through its MCP server: discover APIs, inspect clusters, and perform deliberate administration.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 2 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Start with proxmox_get for /version, /nodes, or /access/permissions.

    line 32
  2. Very specific:

    Use proxmox_request only with a concrete endpoint path and explicit HTTP method.

    line 36
  3. Distinctive:

    Use HTTPS and a trusted CA.

    line 46
  4. Distinctive:

    Run the MCP project's full suite before publishing a local change:

    line 53

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Proxmox VE MCP

This skill carries the audited Proxmox MCP source as a Git submodule at mcp/. Initialize it after cloning this repository:

git submodule update --init --recursive skills/proxmox-ve/mcp
cd skills/proxmox-ve/mcp
bun install --frozen-lockfile
bun run build

Copy mcp.json.example to your user-scoped skill directory as mcp.json, replacing only the absolute installation paths. Keep PROXMOX_URL, token ID, token secret, and CA file in an ignored, mode-0600 environment file; never put them in this skill, MCP configuration, prompts, or tool arguments.

Safe workflow

  1. Start with proxmox_get for /version, /nodes, or /access/permissions.
  2. For unfamiliar operations, use proxmox_search_endpoints and then proxmox_describe_endpoint before constructing the request.
  3. Use proxmox_request only with a concrete endpoint path and explicit HTTP method. POST, PUT, and DELETE can change the cluster.
  4. A returned UPID only confirms submission. Read /nodes/{node}/tasks/{upid}/status and /log before reporting completion.

File and network boundaries

  • Proxmox API tokens enforce authorization; this skill never widens their ACLs.
  • Upload and download paths must be under PROXMOX_FILE_ROOTS. Do not place credentials, private keys, or unrelated user files there.
  • Use HTTPS and a trusted CA. Remote cleartext HTTP is rejected unless the MCP configuration explicitly opts in.
  • API catalog descriptions are advisory. The target PVE server remains authoritative for accepted parameters and permissions.

Verification

Run the MCP project's full suite before publishing a local change:

cd skills/proxmox-ve/mcp
bun run typecheck
bun test
bun run build

Then start a fresh agent session, load this skill explicitly, and make read-only /version and /nodes calls before any administration.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free