clarify-first
✓Use this skill when a request is ambiguous/underspecified, contains conflicts, or has high-impact consequences. The agent must do a quick risk triage (low/medium/high). For low-risk work, proceed with explicit assumptions and reversible steps. For medium/high-risk or conflicting requirements, pause execution, summarize context, list uncertainties, ask targeted clarification questions (with choices), and get confirmation before irreversible actions (writing files, running commands, deleting data, deploying, messaging, spending money). If a better solution exists, propose it and ask the user to choose.
Installation
SKILL.md
This skill prevents “guess-and-run”. It is a meta workflow skill: when requirements are unclear or conflicting, the agent must align with the user before acting.
Do not activate (or keep it minimal) when the request is already precise, low-risk, and has clear acceptance criteria.
Classify the next action as low / medium / high risk.
Use this skill when a request is ambiguous/underspecified, contains conflicts, or has high-impact consequences. The agent must do a quick risk triage (low/medium/high). For low-risk work, proceed with explicit assumptions and reversible steps. For medium/high-risk or conflicting requirements, pause execution, summarize context, list uncertainties, ask targeted clarification questions (with choices), and get confirmation before irreversible actions (writing files, running commands, deleting data, deploying, messaging, spending money). If a better solution exists, propose it and ask the user to choose. Source: dmiyding/clarify-first-skills.
Facts (cite-ready)
Stable fields and commands for AI/search citations.
- Install command
npx skills add https://github.com/dmiyding/clarify-first-skills --skill clarify-first- Category
- {}Data Analysis
- Verified
- ✓
- First Seen
- 2026-02-06
- Updated
- 2026-02-18
Quick answers
What is clarify-first?
Use this skill when a request is ambiguous/underspecified, contains conflicts, or has high-impact consequences. The agent must do a quick risk triage (low/medium/high). For low-risk work, proceed with explicit assumptions and reversible steps. For medium/high-risk or conflicting requirements, pause execution, summarize context, list uncertainties, ask targeted clarification questions (with choices), and get confirmation before irreversible actions (writing files, running commands, deleting data, deploying, messaging, spending money). If a better solution exists, propose it and ask the user to choose. Source: dmiyding/clarify-first-skills.
How do I install clarify-first?
Open your terminal or command line tool (Terminal, iTerm, Windows Terminal, etc.) Copy and run this command: npx skills add https://github.com/dmiyding/clarify-first-skills --skill clarify-first Once installed, the skill will be automatically configured in your AI coding environment and ready to use in Claude Code or Cursor
Where is the source repository?
https://github.com/dmiyding/clarify-first-skills
Details
- Category
- {}Data Analysis
- Source
- skills.sh
- First Seen
- 2026-02-06