genie: Instructions file for Claude Code

CLAUDE.md

genie CLAUDE.md is an instructions file for Claude Code from automagik-dev/genie. It costs 11,942 tokens per session, scanned B, original, MIT.

A set of instructions for Claude Code, an AI coding assistant, when working on the Genie CLI project and its documentation. It explains the project’s checks, build commands, tests, and how its documentation is shared with another repository.

In plain words
What is it for?
It helps run type checks, linting, dead-code checks, builds, and tests; edit Genie documentation; and commit and push documentation changes to the shared docs repository.
Why use it?
It gives the coding assistant the project’s working rules in one place, reducing mistakes when changing code or documentation. It also clarifies that the docs folder links to a separate shared repository.

Instructions file for Claude Code

Written for Claude Code: SessionStart hook event. Also seen: reads .claude/ paths; mentions .claude-plugin; mentions subagents.

This is automagik-dev/genie's own configuration. It tells Claude Code how to work on genie itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything genie configures →

Reuse

Borrowing it

Nothing to install: this file belongs to automagik-dev/genie. 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/automagik-dev/genie/main/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/automagik-dev/genie

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.

agentmods badge for genie CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/automagik-dev/genie/claude-md/github.svg)](https://agentmods.dev/instructions/automagik-dev/genie/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/automagik-dev/genie/claude-md"><img src="https://agentmods.dev/badge/instructions/automagik-dev/genie/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.

agentmods 80×15 button for genie CLAUDE.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/automagik-dev/genie/claude-md"><img src="https://agentmods.dev/badge/instructions/automagik-dev/genie/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 11,942 This file is loaded in full into every session.
When invoked 11,942 The same file — it is already loaded in full.
Security scan B 1 finding. 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.11942 $0.11942
Opus 5.5 $0.04777 $0.04777
Sonnet 5.5 $0.02388 $0.02388
Haiku 4.5 $0.01194 $0.01194

Measured 12d ago against content hash 077c88339344, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade B, and why

genie CLAUDE.md scanned grade B with 1 finding 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 12d 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.

Asks for rootmediumPrivilege escalation

A mod that escalates privileges can change anything on the machine, not only the project.

- **`genie doctor` prints ONE `mode drift` line, never one per drifted entry** — `checkWorktreeModes` (`src/genie-commands/doctor-modes.ts`) returns a single aggregated result naming at most five entries in the fixed ord
CLAUDE.md · 279 lines

How it starts

The opening of the file, as written. The whole thing — 279 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Genie CLI

Claude Code overlay: read and follow the canonical shared repository contract in AGENTS.md first. This file adds Claude-specific command and operational detail; shared rules belong in AGENTS.md.

Commands

bun run check        # Full gate: typecheck + lint + dead-code + test
bun run build        # Bundle to dist/genie.js (bun target, minified, single file)
bun run typecheck    # tsc --noEmit
bun run lint         # biome check .
bun run dead-code    # bunx knip (has pre-existing false positives for biome/commitlint/husky)
bun test             # All tests
bun test src/lib/v5/task-state.test.ts  # Single file

Docs

docs/ is a symlink to .docs-vendor/genie/ where .docs-vendor is a git submodule of automagik-dev/docs (Mintlify, public site at automagik.dev). Engineers see and edit docs/ as if it were a regular subfolder of the genie repo — the submodule machinery is mostly invisible.

  • Operator-facing pages (e.g., docs/installation.mdx, docs/security/key-rotation.mdx, docs/incident-response/canisterworm.mdx) appear on the public Mintlify site at automagik.dev/genie/....
  • Engineering-internal pages live under docs/_internal/ (architecture deep-dives, observability internals, agent-frontmatter contracts, CLI reference dumps, spawn-flow runbooks, detector specs). These are excluded from the public Mintlify build via **/_internal/ in automagik-dev/docs/.mintignore — visible inside the genie repo, hidden from public docs.

Workflow when editing docs:

# Make changes (the symlink follows into .docs-vendor/genie/)
$EDITOR docs/installation.mdx

# Commit + push the docs change to automagik-dev/docs
cd .docs-vendor
git checkout -b feat/<topic>
git add genie/installation.mdx
git commit -m "docs(genie): ..."
git push -u origin feat/<topic>
gh pr create --base main

# After the docs PR merges, bump the genie superproject pointer
cd ..   # back to genie repo root
git submodule update --remote .docs-vendor
git add .docs-vendor
git commit -m "chore: bump .docs-vendor to docs main"

Read the full file on GitHub · 279 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.

  1. 12d ago Changed · +844 tokens per session 077c88339344
  2. 22d ago Changed · +3 lines · +1,071 tokens per session e99bf8dc8ac8
  3. 25d ago Changed · +6 tokens per session 5641c8f5c5fe
  4. 28d ago Changed · +99 tokens per session e9272b96d6bb
  5. 1mo ago Changed · -18 lines · -1,766 tokens per session 1ddcb6ef566d
  6. 1mo ago Changed · -3 tokens per session a349765ce392
  7. 1mo ago Changed · +40 lines · +941 tokens per session scan A → B 2fd889753cde
  8. 1mo ago Changed · +17 lines · +4,549 tokens per session 7a5e664457fb
  9. 1mo ago Changed · -7 lines · +207 tokens per session 1ccab4c30ff7
  10. 1mo ago First seen · 244 lines · 5,994 tokens per session scan A bcbe91457a57

Subscribe to this mod's changes

genie CLAUDE.md is an instructions file published in the GitHub repository automagik-dev/genie (345 stars, last pushed today), licensed MIT. It adds 11,942 tokens to every session, about $0.0478 per session on Opus 5.5. A static security scan graded it B with 1 finding (asks for root). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other instructions, from other repositories