agentmods.dev

Homunculus: Instructions file for Claude Code

CLAUDE.md

Homunculus CLAUDE.md is an instructions file for Claude Code from JavanC/Homunculus. It costs 763 tokens per session, scanned A, original, MIT.

A development guide for Homunculus, a Node.js and Claude CLI project with scripts, commands, templates, and documentation.

In plain words
What is it for?
Use it when changing or releasing Homunculus to follow its project structure and release checklist, including versioning, publishing, and documentation updates.
Why use it?
It prevents releases from missing required updates across code, documentation, translations, version numbers, packages, and GitHub releases.

Instructions file for Claude Code

Written for Claude Code: the file is CLAUDE.md. Also seen: reads .claude/ paths; mentions CLAUDE.md; mentions Codex.

This is JavanC/Homunculus's own configuration. It tells Claude Code how to work on Homunculus 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 Homunculus configures →

not rated 15repo changed 12d ago A scan 763 tokens original MIT
View source ↗ JavanC/Homunculus
Reuse

Borrowing it

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

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 Homunculus CLAUDE.md

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

Your own site · 80×15
<a href="https://agentmods.dev/instructions/javanc/homunculus/claude-md"><img src="https://agentmods.dev/badge/instructions/javanc/homunculus/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 763 This file is loaded in full into every session.
When invoked 763 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00763 $0.00763
Opus 5.5 $0.00305 $0.00305
Sonnet 5.5 $0.00153 $0.00153
Haiku 4.5 $0.00076 $0.00076

Measured 12d ago against content hash 4030d1fd0232, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

Homunculus 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 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.

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 · 54 lines

How it starts

The opening of the file, as written. The whole thing — 54 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Homunculus — Development Guide

Two repos

  • This checkout is the private dev repo (the maintainer's private repo, remote origin). Work and commit here.
  • The public repo (JavanC/Homunculus) is updated only by scripts/release-public.sh, which copies allow-listed paths, runs the tests and a privacy scan, and commits in a separate checkout. Push it by hand after reviewing the diff.
  • Private-only paths (never published): purpose-tree.yaml, docs/plans/, docs/purpose-tree-project.md, private/, scripts/release-public.sh.

Release Checklist

Skill changes (skills/) — the normal case

Users install skills from the public repo (npx skills add JavanC/Homunculus), so a pushed public commit is the release. No version bump, no npm.

  1. Code changes — fix it here, never only in an installed copy of a skill: a reinstall overwrites the installed copy and drops the fix.
  2. Tests — node --test skills/*/tests/*.test.cjs green.
  3. Docs — if behavior a user sees changed, update README.md and docs/zh-TW/README.md, docs/zh-CN/README.md, including a ### skills-YYYY-MM-DD changelog entry (the release notes are read from it).
  4. Review the diff since the last release (Codex first, else a fresh-context Claude agent) and settle every finding.
  5. git commit + push (dev), then scripts/release-public.sh (commits and tags skills-YYYY-MM-DD), review the diff, then run the two commands it prints: git push --follow-tags and gh release create (users watch Releases).
  6. Reinstall globally — npx -y skills add JavanC/Homunculus -g -y; if the global skills folder is version-controlled, commit it together with its .skill-lock.json.

Legacy npm CLI (bin/, core/, commands/, templates/) — frozen

Only for a fix to existing CLI users. After steps 1–5 above: bump package.json (semver), npm publish from repo root, gh release create vX.Y.Z with release notes.

Project Structure

  • skills/ — purpose-tree, purpose-sync, purpose-review; each has SKILL.md, scripts/*.cjs, tests/
  • bin/ — legacy CLI entry points (init, upgrade, night)
  • core/ — legacy evolution scripts copied to user projects during init
  • commands/ — legacy slash commands copied to .claude/commands/ during init
  • templates/ — legacy CLAUDE.md, architecture.yaml, rules templates for init
  • docs/ — Long-form documentation
  • examples/reference/ — Real-world reference implementation from the author's system

Read the full file on GitHub · 54 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 · +3 lines · +87 tokens per session 4030d1fd0232
  2. 15d ago Changed · +11 lines · +174 tokens per session db0fec6e8067
  3. 20d ago Changed · +9 lines · +141 tokens per session 2a41e518cdf8
  4. 2mo ago First seen · 31 lines · 361 tokens per session scan A 3d9f4218400e

Subscribe to this mod's changes

Homunculus CLAUDE.md is an instructions file published in the GitHub repository JavanC/Homunculus (15 stars, last pushed 13d ago), licensed MIT. It adds 763 tokens to every session, about $0.0031 per session on Opus 5.5. 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 2026-08-30.

Related

Homunculus CLAUDE.md compared with instructions from other repositories

Homunculus CLAUDE.md next to 6 instructions from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
Homunculus CLAUDE.mdJavanC/HomunculusInstructions file1513d agoMITA763Claude Code
vscode buildNext.instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA9,719GitHub Copilot
next.js AGENTS.mdvercel/next.jsInstructions file143ktodayMITA7,461Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA5,685GitHub Copilot
langchain AGENTS.mdlangchain-ai/langchainInstructions file149ktodayMITA4,680Codex, OpenCode
vscode copilot-instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA3,093GitHub Copilot
vscode model-prompts.instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA2,022GitHub Copilot