What is homelab_mcp CLAUDE.md?

A set of instructions for developing Homelab MCP, a server that gives an AI access to monitoring and management tools for home servers and services. It covers the project’s technology, authentication, and coding conventions.

Use it when changing the server, its 85 tools, authentication, service integrations, monitoring behavior, or carefully limited management actions.

It keeps changes consistent with the project’s security rules and summary-first design, where an AI can get a useful answer without writing specialist queries or making many tool calls.

How jo-nike/homelab_mcp configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to jo-nike/homelab_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/jo-nike/homelab_mcp/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/jo-nike/homelab_mcp

Made for: Claude Code.

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/jo-nike/homelab_mcp/claude-md/github.svg)](https://agentmods.dev/instructions/jo-nike/homelab_mcp/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/jo-nike/homelab_mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/jo-nike/homelab_mcp/claude-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/jo-nike/homelab_mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/jo-nike/homelab_mcp/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 2,921 This file is loaded in full into every session.

When invoked 2,921 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	$0.02921	$0.02921
Opus 5	$0.01460	$0.01460
Sonnet 5	$0.00584	$0.00584
Haiku 4.5	$0.00292	$0.00292

Measured 9d ago against content hash 89d503757762, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

homelab_mcp CLAUDE.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.

CLAUDE.md · 102 lines

How it starts

The opening of the file, as written. The whole thing — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Project

Homelab MCP

A FastMCP server that unifies the homelab infrastructure behind a single MCP endpoint. 85 tools across 29 upstream services (Prometheus, Loki, Proxmox, Portainer/Docker, Plex, Sonarr/Radarr, Overseerr, Transmission, Synology, PBS, Backblaze, Technitium DNS, Gitea, WireGuard, Scanopy, Vikunja, NPM, Prowlarr, CrowdSec, MySpeed, Healthchecks, Tautulli, LiteLLM, llama-server, SearXNG, OwnTracks, ntfy, internet-health exporter, and cross-service aggregations). Designed primarily for local LLMs (Gemma-class and similar) with summary-first tool design: every tool returns a complete, useful answer without requiring the model to compose queries.

Core Value: Any LLM, local or cloud, can answer questions about the homelab's current state by calling a single tool, without needing domain-specific query languages or multi-step reasoning.

Constraints

Tool design: Summary-first. Every tool group needs pre-built tools that return complete answers; raw query tools (PromQL, LogQL) are secondary.
Read-mostly: The server is primarily monitoring/knowledge. A small, deliberate set of write tools exists (restart_container, safe_restart_container, create_vikunja_task, update_vikunja_task, move_vikunja_task, create_vikunja_comment, create_task_from_alert, overseerr_approve_request, overseerr_decline_request, apply_image_update, send_ntfy_notification). Each is audit-logged to Loki via lib/audit.py, carries readOnlyHint: False (and destructiveHint: True where applicable), and supports dry_run=True; new write tools must follow all three rules. Every write is audit-logged to Loki (best-effort via lib/audit.py). refresh_registries/refresh_docs also mutate server state and are audit-logged too, but carry no dry_run (reloading a cache has no preview) and are marked idempotentHint: True.
Consumer contract: Crow's Nest Watch/Harbor pollers consume get_sonarr_status, get_radarr_status, get_overseerr_requests, get_transmission_torrents and related payloads. Changes to existing tool payload shapes must be additive only; never rename or remove existing fields.
Transport: stdio by default; MCP_TRANSPORT=streamable-http (or http/sse) serves over HTTP via uvicorn on MCP_PORT (8000 default, 5774 in Docker/prod).
Self-signed certs: Proxmox, PBS, Portainer, and Synology clients use verify=False. Do not widen this to other services.
Auth patterns: Per-service (API keys, session logins, token+dynamic-URL). Session-based services go through lib/auth.SessionAuthManager with a per-service LoginStrategy.
Conditional registration: Each tool module's register(mcp) returns early when its credentials are missing from the environment.

Read the full file on GitHub · 102 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 · 102 lines · 2,921 tokens per session scan A 89d503757762

Subscribe to this mod's changes

homelab_mcp CLAUDE.md is an instructions file published in the GitHub repository jo-nike/homelab_mcp (0 stars, last pushed 10d ago), licensed MIT. It adds 2,921 tokens to every session, about $0.0146 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.

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

homelab_mcp: Instructions file for Claude Code