
clarify
by corca-ai
SKILL.md
name: clarify description: Transform vague or ambiguous requirements into precise, actionable specifications through structured questioning. Use when user says "clarify", "refine requirements", or when requirements are unclear, incomplete, or open to multiple interpretations. Always use before implementing any ambiguous feature request. allowed-tools:
- AskUserQuestion
- Write
- Read
Clarify
Quick Start
- Capture: Record the original requirement verbatim
- Clarify: Use
AskUserQuestionto resolve ambiguities iteratively - Compare: Present Before/After comparison
- Save (optional): Offer to save clarified spec to file
Protocol
Phase 1: Capture Original Requirement
Record the original requirement exactly as stated:
## Original Requirement
"{user's original request verbatim}"
Identify ambiguities using the categories below (see Ambiguity Categories).
Phase 2: Iterative Clarification
Use AskUserQuestion tool to resolve each ambiguity. Continue until ALL aspects are clear.
Question Design Principles:
- Specific over general: Ask about concrete details, not abstract preferences
- Options over open-ended: Provide 2-4 choices (recognition > recall)
- One concern at a time: Avoid bundling multiple questions
- Neutral framing: Present options without bias
Loop Structure:
while ambiguities_remain:
identify_most_critical_ambiguity()
ask_clarifying_question() # Use AskUserQuestion tool
update_requirement_understanding()
check_for_new_ambiguities()
Phase 3: Before/After Comparison
After clarification is complete, present the transformation:
## Requirement Clarification Summary
### Before (Original)
"{original request verbatim}"
### After (Clarified)
**Goal**: [precise description of what user wants]
**Reason**: [the ultimate purpose or jobs to be done]
**Scope**: [what's included and excluded]
**Constraints**: [limitations, requirements, preferences]
**Success Criteria**: [how to know when done]
**Decisions Made**:
| Question | Decision |
|----------|----------|
| [ambiguity 1] | [chosen option] |
| [ambiguity 2] | [chosen option] |
Phase 4: Save Option
Use AskUserQuestion to ask if the user wants to save the clarified requirement:
- Question: "Save this requirement specification to a file?"
- Options: "Yes, save to file" (Save to requirements/ directory) / "No, proceed" (Continue without saving)
If saving:
- Default location:
requirements/or project-appropriate directory - Filename: descriptive, based on requirement topic (e.g.,
auth-feature-requirements.md) - Format: Markdown with Before/After structure
Ambiguity Categories
Common types to probe:
| Category | Example Questions |
|---|---|
| Scope | What's included? What's explicitly out? |
| Behavior | Edge cases? Error scenarios? |
| Interface | Who/what interacts? How? |
| Data | Inputs? Outputs? Format? |
| Constraints | Performance? Compatibility? |
| Priority | Must-have vs nice-to-have? |
| Reason | Why are you doing this? What are jobs to be done? |
| Success | How to verify we're taking the right steps? |
Examples
Example 1: Vague Feature Request
Original: "Add a login feature"
Clarifying questions (via AskUserQuestion):
- Authentication method? → Username/Password
- Registration included? → Yes, self-signup
- Session duration? → 24 hours
- Password requirements? → Min 8 chars, mixed case
Clarified:
- Goal: Add username/password login with self-registration
- Reason: Restrict access to authorized users only
- Scope: Login, logout, registration, password reset
- Constraints: 24h session, bcrypt, rate limit 5 attempts
- Success Criteria: User can register, login, logout, reset password
Example 2: Bug Report
Original: "The export is broken"
Clarifying questions:
- Which export? → CSV
- What happens? → Empty file
- When did it start? → After v2.1 update
- Steps to reproduce? → Export any report
Clarified:
- Goal: Fix CSV export producing empty files
- Reason: Users need to export data for external reporting
- Scope: CSV only, other formats work
- Constraints: Regression from v2.1, must not break other exports
- Success Criteria: CSV contains correct data matching UI
Rules
- No assumptions: Ask, don't assume
- Preserve intent: Refine, don't redirect
- Minimal questions: Only what's needed
- Respect answers: Accept user decisions
- Track changes: Always show before/after
スコア
総合スコア
リポジトリの品質指標に基づく評価
SKILL.mdファイルが含まれている
ライセンスが設定されている
100文字以上の説明がある
GitHub Stars 100以上
3ヶ月以内に更新がある
10回以上フォークされている
オープンIssueが50未満
プログラミング言語が設定されている
1つ以上のタグが設定されている
レビュー
レビュー機能は近日公開予定です