Repository inventory

haroutkhac/code-comprehension-quiz

Skills indexed from this repository, with install-style signals scoped to the repo.
1 skills0 GitHub stars0 weekly installsGitHubOwner profile

Overview

This skill generates adaptive multiple-choice quizzes to test a developer's understanding of recent code changes or explanations. It only runs when the user explicitly asks (e.g., "quiz me", "test my understanding", or "/quiz") and never auto-triggers. The quiz adapts length and difficulty to the scope and complexity of the work just done.

How this skill works

On explicit request the skill inspects the most recent change and conversation context, then generates 2–5 focused multiple-choice questions that probe tradeoffs, failure modes, alternatives, implications, root causes, and edge cases. It presents questions one at a time, records answers, scores the quiz, and provides a concise review explaining correct answers and misconceptions. After a quiz finishes it waits for another explicit request before running again.

When to use it

  • After completing a coding task when you want to check comprehension
  • Following a code review or explanation to verify attention to tradeoffs
  • When preparing for onboarding or handoff to ensure understanding
  • Before merging a complex change to surface overlooked implications
  • When studying recent debugging work to confirm root-cause insight

Best practices

  • Trigger the quiz only with explicit commands like "quiz me" or "/quiz"
  • Provide the most recent diff or explanation in the conversation for focused questions
  • Answer questions in sequence; the skill marks unanswered questions as skipped
  • Use the review summary to learn from missed questions and request a deeper dive if score < 80%
  • Limit each quiz to the most recent work so questions remain answerable from context

Example use cases

  • After refactoring a component, ask "quiz me" to test understanding of tradeoffs
  • Post-bugfix, use "/quiz" to validate knowledge of the root cause and failure modes
  • During code review, request a quiz to ensure reviewers noticed implications
  • For onboarding, test new team members on recent architectural decisions

FAQ

No. The skill never auto-triggers. It runs only when you explicitly ask with phrases like "quiz me" or "/quiz".

How many questions will I get?

Question count adapts to complexity: 2 for low, 3 for medium, and 4–5 for high complexity changes.

How are correct answer positions chosen?

Positions are randomized per question using a deterministic formula tied to the question number and option count to avoid simple cycles.

1 skills

More from this maintainer
Other repositories and skills published under the same GitHub owner.
Skills library
Jump back to the full directory or explore grouped topics.
Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational