A command that opens a software-development challenge and explains what must be built. It reads the challenge specification, acceptance checklist, and relevant design files before asking how you plan to proceed.

Use it to review endpoints, request and response formats, status codes, acceptance criteria, design test IDs, dependencies, and challenge-specific traps.

It gathers the contract and evaluation rules in one orientation, so you can understand the task and common pitfalls before coding. It also identifies missing prerequisites and unclear requirements.

challenge — Command by micheltlutz

Install

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.

agentmods

npx agentmods add commands/micheltlutz/dev-challenge/challenge

Clone the repo

git clone --depth 1 https://github.com/micheltlutz/dev-challenge

Made for: Claude Code.

Per session 17 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 434 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. Scan, not verified.

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	$0.00017	$0.00434
Opus 5	$0.00009	$0.00217
Sonnet 5	$0.00003	$0.00087
Haiku 4.5	$0.00002	$0.00043

Measured 2d ago against content hash c1793f49273a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

challenge 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 2d 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/commands/challenge.md · 51 lines

What it actually says

Start work on challenge $1.

Load the context

Read specs/challenges/$1/spec.md. If the directory does not exist, run python harness/cli.py --list and suggest the closest match.
Read specs/challenges/$1/acceptance.md.
If the frontmatter has mockups: or figma_node:, also read specs/design/style-guide.md and note the testids: list.
Check depends_on: — if a prerequisite has not been done, say so before starting.

Then

Give the person a short orientation, not a wall of text:

What they are building, in two or three sentences.
The contract: endpoints, request and response shapes, status codes.
The acceptance criteria as a numbered checklist, with IDs.
Anything the spec leaves genuinely ambiguous.
The one trap most people hit on this specific challenge.

Then ask what their approach is, and work from there.

Posture

Default: mentor. Plan with them, review what they write, explain failing criteria. Do not write the implementation. Scaffolding, config and boilerplate are fine to write outright — nobody learns from a tsconfig.json.

If $ARGUMENTS contains --solve, or they ask directly, switch to solver mode: say you have switched, then implement it properly and walk them through the reasoning as you go.

See the challenge-workflow skill for the full loop.

Verifying

python harness/cli.py --base-url <their-url> --challenge $1

Mobile challenges have no automated runner — say so plainly rather than implying coverage that does not exist.

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.

2d ago First seen · 51 lines · 17 tokens per session scan A c1793f49273a

Subscribe to this mod's changes

challenge is a command published in the GitHub repository micheltlutz/dev-challenge (32 stars, last pushed 15d ago), licensed MIT. It adds 17 tokens to every session and 434 once invoked, about $0.0001 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.

Backend and APIs Code review Testing challenge development fastapi front-end front-end-challenge

specify

Create or update the feature specification from a natural language feature description.

github/spec-kit · 13 tokens

speckit.git.validate

Validate current branch follows feature branch naming conventions.

github/spec-kit · 8 tokens

speckit.assess.intake

Capture and normalize a raw idea (text, URL, ticket, or codebase pointer) into an intake note.

github/spec-kit · 22 tokens

speckit.bug.assess

Assess a bug report (pasted text or URL) against the codebase and produce an assessment with possible remediation.

github/spec-kit · 22 tokens

checkpoint

Create a checkpoint by staging the changes you have made using git. usage: git [-v | --version] [-h | --help] [-C ] [-c = ] [--exec-path[= ]] [--html-path] [--man-path] [--info-path] [-p | --paginate | -P | --no-pager] [--no-replace-objects] [--bare] [--git-dir= ] [--work-tree= ] [--namespace= ] [--super-prefix= ]…

TracecatHQ/tracecat · 0 tokens

os-meeting

快速创建和管理AI团队会议.

CronusL-1141/AI-company · 11 tokens