Clarify first

Skill Rookage/agent-discipline/skills/clarify-first

Practical bilingual skills that make AI coding agents more disciplined, testable, and useful in real engineering work.

Install

npx -y skills add Rookage/agent-discipline --skill clarify-first

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Use before starting unclear product, feature, architecture, workflow, or business tasks. Clarifies requirements, terminology, scope, edge cases, and decisions; updates CONTEXT.md and ADRs only when useful. Trigger when user says ask first, clarify, plan, PRD, do not rush, or discuss before coding.

SKILL.md

1.4 KB, as published. Nobody here has run it

Clarify First

中文核心：先问清楚，再开工。能从代码判断的，不问用户。

Workflow

Read the local context first.
- Look for README, existing docs, CONTEXT.md, CONTEXT-MAP.md, docs/adr/, and nearby code.
Ask one useful question at a time.
- Do not dump a questionnaire.
- Include a recommended answer when possible.
- If code answers the question, use the code.
Sharpen language.
- Identify overloaded terms.
- Propose canonical names.
- Use the project's existing vocabulary when present.
Stress-test with concrete scenarios.
- Happy path, empty state, permission failure, duplicate action, retry, rollback, edge cases.
Capture durable knowledge.
- CONTEXT.md: domain terms only, not implementation notes.
- ADR: only for decisions that are hard to reverse, surprising, and based on real tradeoffs.
Move to execution.
- Once the main branches are resolved, produce a short plan and start.

Style

One question beats ten.
Do not make the user manage the agent.
Be precise without being verbose.