What is mcp AGENTS.md?

Repository instructions for the GrowthKit MCP worker, a Cloudflare Worker that provides an MCP server. MCP is a standard way for AI clients to discover and call tools; the instructions document the codebase's rules, known facts, and verification practices.

Use it when modifying the GrowthKit MCP worker, its AGENTS.md and CLAUDE.md guidance, Cloudflare or Supabase integration, MCP endpoints, OAuth, workflows, or golden-master tests.

It reduces mistakes when changing a public repository whose MCP protocol, deployment setup, and regression checks are implemented manually.

How growthkit-tools/mcp configures Claude Code and Codex

Reuse

Borrowing it

Nothing to install: this file belongs to growthkit-tools/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/growthkit-tools/mcp/main/AGENTS.md

Clone the repo

git clone --depth 1 https://github.com/growthkit-tools/mcp

Made for: Claude Code, 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/growthkit-tools/mcp/agents-md/github.svg)](https://agentmods.dev/instructions/growthkit-tools/mcp/agents-md)

Your own site

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

Per session 22,644 This file is loaded in full into every session.

When invoked 22,644 The same file — it is already loaded in full.

Security scan B 2 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	$0.22644	$0.22644
Opus 5.5	$0.09058	$0.09058
Sonnet 5.5	$0.04529	$0.04529
Haiku 4.5	$0.02264	$0.02264

Measured 1mo ago against content hash 164cb47537ac, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade B, and why

mcp AGENTS.md scanned grade B with 2 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 1mo 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.

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

**Die Regel wurde am 21.08. aus `~/.claude/settings.json` entfernt — nicht

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

| zsh-Schleifenvariable `path` | „33 von 33 Abweichungen" | `path` ist an `PATH` gebunden; `curl` war nach der ersten Zeile nicht mehr auffindbar |

AGENTS.md · 1,257 lines

How it starts

The opening of the file, as written. The whole thing — 1,257 lines — stays where its author put it; the contents beside it link to each section on GitHub.

GrowthKit MCP Worker — Agent Quick Reference

Ablage: Repo-Root als AGENTS.md. Danach git rm CLAUDE.md && ln -s AGENTS.md CLAUDE.md — eine Quelle, zwei Namen. Zwei getrennte Guide-Dateien driften auseinander.

Prinzip: Jede Zeile geht auf einen beobachteten Fehler oder eine reale Invariante zurück. Neue Zeilen nur bei echtem Anlass. Zeilen ohne Anlass sind Rauschen.

⚠️ TODO = gegen das Repo verifizieren. 🔵 ENTSCHEIDUNG = Chris muss bestätigen.

Wissen

Cloudflare Worker, serviert den GrowthKit MCP-Server auf mcp.growthkit.tools.

Ein hand-rolled File: index.js, ~4200 Zeilen (Stand 24.08.2026). MCP JSON-RPC manuell implementiert, kein SDK. tools/list, tools/call, resources/list, resources/read, Prompts und OAuth sind alle von Hand gebaut. Es gibt keine Bibliothek, die Fehler abfängt.
Konsumenten: Claude (Desktop/Web/Code), ChatGPT, MCP-Registries. Regressionen brechen still — kein Alert, nur sinkende Nutzung. Deshalb ist der Golden Master nicht optional.
Backend: Supabase-Projekt (EU Frankfurt) über Edge Functions, Custom Domain api.growthkit.tools proxied /functions/v1/*. Project-Ref steht in wrangler.toml / den Worker-Secrets, nicht hier.
Auth: Zwei Auflösungswege, unterschieden allein am Token-Präfix (index.js ~1321). Die Entscheidung fällt vor jedem Backend-Aufruf.
- Bearer beginnt mit gk_ → direkter API-Token-Pfad: resolve_user_token gegen user_api_tokens; die RPC hasht selbst und prüft is_active. Kein OAuth, kein Browser — das ist der Pfad für Tests, CI und Automatisierung. Zwei bewusste Nicht-Handlungen: last_used_at bleibt ungeschrieben (resolve_user_token schreibt nicht, das tut nur use_api_token), Tokens sehen über diesen Weg also unbenutzt aus; und isDemo ist hier immer false — user_api_tokens hat kein is_demo, Demo-Sessions laufen ausschließlich über OAuth.
- Alles andere → OAuth-Pfad wie bisher: gegen oauth_tokens.access_token mit Ablaufprüfung; das gk_-Token liegt dort als user_token.

Read the full file on GitHub · 1,257 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.

1mo ago Changed · +6 lines · +102 tokens per session 164cb47537ac
1mo ago Changed · +26 lines · +387 tokens per session d75be57a1f26
2mo ago Changed · +68 lines · +1,292 tokens per session df133406aeac
2mo ago First seen · 1,157 lines · 20,863 tokens per session scan B 06855886fc72

Subscribe to this mod's changes

mcp AGENTS.md is an instructions file published in the GitHub repository growthkit-tools/mcp (0 stars, last pushed 1mo ago), licensed MIT. It adds 22,644 tokens to every session, about $0.0906 per session on Opus 5.5. A static security scan graded it B with 2 findings (reads agent configuration directories, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

mcp AGENTS.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
mcp AGENTS.mdgrowthkit-tools/mcp	Instructions file	0	1mo ago	MIT	B	22,644	Claude Code, Codex, OpenCode
vscode buildNext.instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	9,719	GitHub Copilot
next.js AGENTS.mdvercel/next.js	Instructions file	143k	today	MIT	A	7,461	Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	5,685	GitHub Copilot
langchain AGENTS.mdlangchain-ai/langchain	Instructions file	149k	today	MIT	A	4,680	Codex, OpenCode
vscode copilot-instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	3,093	GitHub Copilot
vscode model-prompts.instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	2,022	GitHub Copilot

mcp: Instructions file for Claude Code