What is proxmox-mcp AGENTS.md?

Repository-specific instructions for proxmox-mcp, a TypeScript server that lets AI tools work with Proxmox VE, a platform for managing virtual machines and containers.

Use them when modifying the server, its Proxmox tools, SSH execution, security code, or build and test workflow.

They tell coding agents how this repository is structured, which safety rules apply, and what checks define a completed change.

How lidless-labs/proxmox-mcp configures Codex and OpenCode

Reuse

Borrowing it

Nothing to install: this file belongs to lidless-labs/proxmox-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/lidless-labs/proxmox-mcp/master/AGENTS.md

Clone the repo

git clone --depth 1 https://github.com/lidless-labs/proxmox-mcp

Made for: Codex, OpenCode.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/lidless-labs/proxmox-mcp/agents-md/github.svg)](https://agentmods.dev/instructions/lidless-labs/proxmox-mcp/agents-md)

Your own site

<a href="https://agentmods.dev/instructions/lidless-labs/proxmox-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/lidless-labs/proxmox-mcp/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/lidless-labs/proxmox-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/lidless-labs/proxmox-mcp/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 1,236 This file is loaded in full into every session.

When invoked 1,236 The same file — it is already loaded in full.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.01236	$0.01236
Opus 5	$0.00618	$0.00618
Sonnet 5	$0.00247	$0.00247
Haiku 4.5	$0.00124	$0.00124

Measured 9d ago against content hash bc2226324b0e, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

proxmox-mcp AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

AGENTS.md · 43 lines

How it starts

The opening of the file, as written. The whole thing — 43 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Repository Guidance

Definition of Done

./scripts/verify

It runs npm run typecheck, npm test, and npm run build in order, with tests against the fake Proxmox. A change is complete only when it passes after your final edit.

Any edit after a passing run invalidates that run. Re-run it.
Report actual results. If anything fails, report the failure verbatim and stop; do not claim success, do not soften the error.
prepublishOnly runs the same chain; the publish payload is only dist, openclaw.plugin.json, README.md, LICENSE.

Project Shape

TypeScript MCP server for Proxmox VE (@solomonneas/proxmox-mcp): 96 tools across read, gated guest-read, safe-write, and destructive tiers, plus in-guest exec over SSH.
Entry points: mcp-server.ts (stdio MCP), index.ts (OpenClaw plugin). Both register tools through the explicit export list in src/tools/index.ts; that list is canonical.
One tool implementation per file under src/tools/. Shared plumbing: src/proxmox-client.ts (API token HTTP client), src/ssh-executor.ts (host pct exec for LXC, direct SSH for QEMU), src/gates.ts (write gates), src/security.ts (secret redaction), src/config.ts (env resolution).
Tests use the in-memory fake in tests/fake-proxmox.ts; per-tool tests live in tests/tools/.

Hard Prohibitions

The configured Proxmox host is a live home-lab hypervisor running production CTs and VMs. Never run write, exec, service-action, or destructive tools against live VMIDs during development or review unless the user explicitly asks for that live operation in this session. Default to the fake (tests/fake-proxmox.ts) and read-only tools.
The five destructive tools are proxmox_destroy_resource, proxmox_cleanup_smoke_resources, proxmox_rollback_snapshot, proxmox_delete_snapshot, proxmox_force_stop_resource. Each requires the full gate chain: assertConfirmedWrite(raw, NAME), then assertDestructive(raw, NAME), then assertEnvFlag on PROXMOX_ENABLE_DESTRUCTIVE, all before any network request. Never weaken, reorder, bypass, or conditionally skip any gate, in code or in tests.
Smoke tests stay confined to the smoke pool: only mcp-smoke-* resources in the mcp-smoke pool. Never point smoke at any other VMID.
A test fails: fix the code or fix a genuinely wrong test. Never delete, skip, loosen assertions on, or mark-flaky a failing test to get green.
This repo has a pre-push hook (content-guard scan, hooks/pre-push via core.hooksPath). Never push with --no-verify. If the hook blocks, report its output verbatim.
Blocked on credentials, env flags, network, or permissions: report the exact blocker and stop. Do not invent workarounds that route around a gate or a hook.

Read the full file on GitHub · 43 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

9d ago First seen · 43 lines · 1,236 tokens per session scan A bc2226324b0e

Subscribe to this mod's changes

proxmox-mcp AGENTS.md is an instructions file published in the GitHub repository lidless-labs/proxmox-mcp (7 stars, last pushed 1mo ago), licensed MIT. It adds 1,236 tokens to every session, about $0.0062 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

homelab mcp mcp-server model-context-protocol proxmox

next.js AGENTS.md

AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,296 tokens

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,153 tokens

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

proxmox-mcp: Instructions file for Codex