Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/micheltlutz/dev-challenge/claude-mdgit clone --depth 1 https://github.com/micheltlutz/dev-challengeWrote 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.
[](https://agentmods.dev/instructions/micheltlutz/dev-challenge/claude-md)<a href="https://agentmods.dev/instructions/micheltlutz/dev-challenge/claude-md"><img src="https://agentmods.dev/badge/instructions/micheltlutz/dev-challenge/claude-md.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00543 | $0.00543 |
| Opus 5 | $0.00271 | $0.00271 |
| Sonnet 5 | $0.00109 | $0.00109 |
| Haiku 4.5 | $0.00054 | $0.00054 |
Grade A, and why
dev-challenge 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 4d 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.
How it starts
The opening of the file, as written. The whole thing — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
@AGENTS.md
Everything above applies. What follows is Claude Code specific.
The one rule to internalise
Read specs/challenges/<id>/spec.md and acceptance.md before answering
anything about what a challenge requires. Never answer from readme_files/
(redirect stubs) or from how such an API usually works.
And decide which job you are doing — mentoring a learner or improving the
lab. The default when someone is working through a challenge is to coach, not
to hand over the implementation. See "First: which job are you doing?" in
AGENTS.md.
Slash commands
| Command | What it does |
|---|---|
/challenge <id> |
Load a challenge's spec, criteria and design tokens, and start work on it |
/verify [id] |
Run the harness and explain failures in terms of the criteria |
/next |
Suggest what to do next from the dependency graph and what currently passes |
/challenge <id> --solve switches to solver mode explicitly.
Subagents
| Agent | Use it for |
|---|---|
challenge-mentor |
Reviewing a learner's work without solving it for them |
spec-auditor |
Checking a spec change against the contract before a PR |
Skills
| Skill | Loads when |
|---|---|
challenge-workflow |
Someone is working through a challenge |
design-fidelity |
Comparing an implemented UI against the tokens and mockups |
spec-authoring |
Adding or editing a challenge in specs/ |
MCP
.mcp.json declares an optional Figma Dev Mode server. With it,
you can pull real node specs instead of inferring layout from a PNG. It needs
the user's own Figma token and is inert without one — the sampled tokens in
specs/design/tokens.json are the fallback, and they are good enough for every
challenge.
Working in this repo
- The venv is at
./venv. Use./venv/bin/python, or activate it. - Before saying you are done, run what CI runs (see Commands in
AGENTS.md). dev-challenge.dbis generated. Changescripts/seed.pyand regenerate; do not hand-edit it.specs/openapi.jsonandspecs/design/tokens.cssare generated. Same rule.
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.
- 4d ago First seen · 58 lines · 543 tokens per session scan A f80bbf99b4aa
dev-challenge CLAUDE.md is an instructions file published in the GitHub repository micheltlutz/dev-challenge (32 stars, last pushed 2d ago), licensed MIT. It adds 543 tokens to every session, about $0.0027 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 2026-08-30.
Other instructions, from other repositories
android-agent CLAUDE.md
Instructions for ghost-in-the-droid/android-agent, covering ghost in the droid — agent tool rules, mcp tool: opencamera, mcp tool: speaktext and general rule.
MobiAI-Core CLAUDE.md
Claude Code instructions for ArisGuimera/MobiAI-Core, covering mobiai — mobile development ai ecosystem, bootstrap and principles.
geolens CLAUDE.md
Claude Code instructions for geolens-io/geolens: This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
freetube CLAUDE.md
Instructions for leshkodev/freetube, covering claude.md, 1. what this project is, 2. non-negotiable constraints, 3. tech stack (locked) and 4. project structure.
AI-System-Design-Consultant CLAUDE.md
Claude Code instructions for deepanshu2711/AI-System-Design-Consultant, covering claude.md, what this is, running it, load-bearing typos — do not "fix" these paths and agent node pattern.
mcpelevator CLAUDE.md
Instructions for pacnpal/mcpelevator, covering commands, architecture, conventions, agent skills and issue tracker.