02-challenge — Claude Code skill: review work and install
02-challenge is a Claude Code skill that checks completed work against an agreed plan and returns a structured report. Its findings help decide whether to ship, revise, or rework the result.
MIT ★ 571 updated 2026-10-29
/plugin marketplace add ai-driven-dev/framework
/plugin install aidd-refine@aidd-frameworkWhat 02-challenge does
The skill compares finished work with the agreed reference and flags errors, omissions, and duplication. It labels each finding as a deal-breaker, a suggestion, or correct, then assigns a confidence score.
Its report follows assets/report-template.md, while references/confidence-rubric.md guides the confidence rating. The skill also looks for ways to simplify the work. It is meant for critical review, not line-by-line style checks or code writing.
What it helps with
- compare finished work with an agreed plan
- spot missing or duplicated work
- separate deal-breakers from suggestions
- judge confidence before deciding whether to ship
Example prompts for the agent
- Compare this work with the agreed plan and write a structured report.
- Find errors, omissions, and duplication in this result, then rate your confidence.
- Review the completed task against the plan and tell me what needs reworking.
How to install 02-challenge
1Plugin marketplace
/plugin marketplace add ai-driven-dev/framework/plugin install aidd-refine@aidd-frameworkThis installs the whole aidd-refine plugin with all its skills
2Skills CLI (Vercel)
npx skills add ai-driven-dev/framework --skill 02-challenge -a claude-code3Manually: put the files in place
~/.claude/skills/02-challenge/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/02-challenge/
1Skills CLI (Vercel)
npx skills add ai-driven-dev/framework --skill 02-challenge -a codex2Manually: put the files in place
.agents/skills/02-challenge/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add ai-driven-dev/framework --skill 02-challenge -a cursor2Manually: put the files in place
.cursor/skills/02-challenge/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install ai-driven-dev/framework plugins/aidd-refine/skills/02-challenge2Manually: put the files in place
.github/skills/02-challenge/SKILL.md1Skills CLI (Vercel)
npx skills add ai-driven-dev/framework --skill 02-challenge -a gemini-cli2Platform CLI
gemini skills install https://github.com/ai-driven-dev/framework --path plugins/aidd-refine/skills/02-challengeManually: put the files in place
.devin/skills/02-challenge/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add ai-driven-dev/framework --skill 02-challenge -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about 02-challenge
How do I install 02-challenge in Claude Code?
Add the marketplace with /plugin marketplace add ai-driven-dev/framework, then install the plugin with /plugin install aidd-refine@aidd-framework.
What does 02-challenge do?
It reviews completed work against an agreed plan, classifies its findings, and scores confidence. The report helps you decide whether to ship, revise, or rework.
How does 02-challenge score confidence?
It uses the rubric in references/confidence-rubric.md. Findings are laid out using assets/report-template.md.
Source text
SKILL.md · MIT
--- name: 02-challenge description: Rethink just-completed work against an agreed plan, classifying findings as deal-breaker, suggestion, or correct, with a confidence score. Use to challenge or critically review recent work. Not for line-by-line style review or writing code. argument-hint: work --- # Challenge Rethink prior work and surface what is wrong, missing, or duplicated. Output a structured report with a confidence score so the user knows whether to ship, iterate, or rework. ## Actions | # | Action | Role | Input | | --- | ----------- | ------------------------------------------------------------- | ------------------------------ | | 01 | `challenge` | Rethink prior work, classify findings, score confidence | the work + agreed reference | Before running an action, read its file in `actions/`, not only the table or assets. ## Transversal rules - Reason from first principles, no logical gaps. - Aim for simplifications. If the work can be smaller, say so. - Fill `assets/report-template.md` verbatim. ## References - `references/confidence-rubric.md`: tiered rubric for the confidence percentage. ## Assets - `assets/report-template.md`: findings report skeleton, filled per run.
© AI-Driven Dev · MIT · ai-driven-dev/framework