change-quiz — Claude Code skill: report and setup
This skill turns a work session into a change report and a merge quiz. Use it when a large diff needs review or you want to check what the change really affects.
npx skills add Neeeophytee/finding-unknowns-skills --skill change-quiz -a claude-codeWhat change-quiz does
change-quiz builds a four-part report: the original problem, changes grouped by intent, affected code paths, and the mental-model updates a maintainer should retain. It covers behavior that the diff may not show, including how the new code interacts with existing paths.
The report is followed by 5 to 8 questions about deviations, edge cases, and interaction effects. Some ask for recall; others ask the user to predict what will happen. For a long session, the report can be delivered as one self-contained HTML page with the quiz at the bottom.
The skill grades one round at a time. Each wrong answer gets the correct explanation and a diagnosis: a gap in the user’s model or a change that is too clever. After two failed rounds, it recommends simplifying or splitting the change instead of continuing to quiz.
What it helps with
- summarize what a coding session changed
- trace new behavior through existing code paths
- test maintainer understanding before a merge
- explain mistakes in edge-case predictions
- decide whether a change should be split
Example prompts for the agent
- Review this session, explain what changed, and prepare the merge quiz.
- Turn this large change into a four-part report, then test me with 5 to 8 questions.
- Ask me about edge cases and interaction effects, grade one round at a time, and explain every miss.
How to install change-quiz
1Skills CLI (Vercel)
npx skills add Neeeophytee/finding-unknowns-skills --skill change-quiz -a claude-code2Manually: put the files in place
~/.claude/skills/change-quiz/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/change-quiz/
1Skills CLI (Vercel)
npx skills add Neeeophytee/finding-unknowns-skills --skill change-quiz -a codex2Manually: put the files in place
.agents/skills/change-quiz/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add Neeeophytee/finding-unknowns-skills --skill change-quiz -a cursor2Manually: put the files in place
.cursor/skills/change-quiz/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install Neeeophytee/finding-unknowns-skills change-quiz2Manually: put the files in place
.github/skills/change-quiz/SKILL.md1Skills CLI (Vercel)
npx skills add Neeeophytee/finding-unknowns-skills --skill change-quiz -a gemini-cli2Platform CLI
gemini skills install https://github.com/Neeeophytee/finding-unknowns-skills --path skills/change-quizManually: put the files in place
.devin/skills/change-quiz/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add Neeeophytee/finding-unknowns-skills --skill change-quiz -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 change-quiz
How do I install change-quiz in Claude Code?
Run npx skills add Neeeophytee/finding-unknowns-skills --skill change-quiz -a claude-code, then use ~/.claude/skills/change-quiz/SKILL.md.
What does change-quiz do?
It creates a report about the session’s context, code changes, affected paths, and behavior shifts, then runs a quiz before merge. The questions focus on deviations and edge cases rather than name recall.
How many questions does the change-quiz quiz contain?
The quiz has 5 to 8 questions. It mixes recall with predictions about changed behavior, especially where edge cases and code-path interactions could surprise a maintainer.
Source text
SKILL.md · MIT
---
name: change-quiz
description: After a working session, produce a report on what changed plus a quiz the user must pass before merging. Use when the user asks "what did we actually do," wants to review a large change, or invokes a quiz before merge.
---
# Change quiz
After a long session the agent has often done more than the user realizes, and a diff only shows surface. Behavior lives in how the change interacts with existing code paths. The user should merge only what they can pass a quiz on.
## Steps
1. Build the report first, in four short sections:
- **Context** — what problem this session set out to solve.
- **What changed** — grouped by intent (feature, fix, refactor), not by file.
- **How it interacts** — the existing code paths the change touches, and what now behaves differently even in files the diff doesn't show.
- **Intuition** — the 2-3 mental-model updates the user should walk away with ("retries are now idempotent because X").
2. For long sessions, offer the report as a single self-contained HTML page with the quiz at the bottom — it reads better than a wall of markdown.
3. Then the quiz: 5-8 questions targeting what would bite an unaware maintainer.
- Mix recall ("what happens to in-flight jobs during deploy now?") with prediction ("if someone calls X with a stale token, what do they see?").
- Weight questions toward deviations, edge cases, and interaction effects — not trivia about names.
4. Grade honestly, one round at a time. For each miss, explain the right answer AND flag it: a miss is either a gap in the user's model or a sign the change is too clever — say which.
5. Pass = merge-ready. Fail = point back to the specific report sections to reread, then offer a fresh variant quiz. Do not soften the bar; the whole point is that unread changes don't ship.
## Guardrails
- The quiz covers the change and its blast radius, not general knowledge.
- If the user can't pass after two rounds, the recommendation is to simplify the change or split it, not to keep quizzing.
- Never mark the user correct out of politeness. A false pass defeats the skill.
© Neeeophytee · MIT · Neeeophytee/finding-unknowns-skills