← スキル一覧に戻る

clarify
by zettalyst
A collection of plugins for Claude Code
⭐ 0🍴 0📅 2026年1月13日
SKILL.md
name: clarify description: Start iterative requirement clarification loop. Use this skill when the user has a vague requirement and wants to clarify it through structured questioning. Transforms unclear requirements into precise, actionable specifications.
Clarify Skill
Transform vague requirements into precise specifications through structured questioning.
Invocation
Parse the input to extract:
- REQUIREMENT: The vague requirement text (required)
- --max-iterations N: Maximum question rounds (default: 3)
Initialization
- Create
.claude/clarify.local.md:
---
active: true
iteration: 1
max_iterations: [MAX_ITERATIONS]
original_requirement: "[REQUIREMENT]"
started_at: "[ISO timestamp]"
---
## Original Requirement
"[REQUIREMENT]"
## Clarification Progress
(Decisions will be recorded as the loop progresses)
- Confirm activation:
Clarify loop activated!
Original Requirement: "[REQUIREMENT]"
Max Questions: [MAX_ITERATIONS]
To cancel: /cancel
Clarification Methodology
You are now in a Clarify Loop. Transform the vague requirement into a precise, actionable specification through iterative questioning.
Your Task
- Analyze the original requirement for ambiguities
- Ask 4 questions in a single
AskUserQuestioncall to resolve multiple ambiguities at once - Record all answers in your understanding
- Check completion: Include "Clarification complete" as the last option of the last question
- Repeat until user confirms completion OR max iterations reached
Question Design Guidelines
- Ask exactly 4 questions per iteration (use all 4 question slots)
- Provide exactly 4 options per question (use all 4 option slots)
- The 4th question's 4th option should be "Clarification complete - proceed with current understanding"
- Use neutral framing without bias toward any option
- Cover different ambiguity categories in each question set
Ambiguity Categories 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? What's the jobs to be done? |
| Success | How to verify completion? |
Completion Rules
When the user selects "Clarification complete" OR you reach max iterations:
- Output the final clarified requirement:
## Requirement Clarification Summary
### Before (Original)
"{original requirement verbatim}"
### After (Clarified)
**Goal**: [precise description]
**Reason**: [the ultimate purpose]
**Scope**: [what's included and excluded]
**Constraints**: [limitations, requirements]
**Success Criteria**: [how to verify]
**Decisions Made**:
| Question | Decision |
|----------|----------|
| [ambiguity 1] | [chosen option] |
| [ambiguity 2] | [chosen option] |
- Then output:
<promise>CLARIFICATION COMPLETE</promise>
CRITICAL: Only output the promise when you have genuinely finished clarification. Do NOT lie to exit the loop.
Now begin by:
- Creating the state file
- Identifying 4 key ambiguities
- Asking 4 clarifying questions (each with 4 options) using a single AskUserQuestion call
スコア
総合スコア
50/100
リポジトリの品質指標に基づく評価
✓SKILL.md
SKILL.mdファイルが含まれている
+20
○LICENSE
ライセンスが設定されている
0/10
○説明文
100文字以上の説明がある
0/10
○人気
GitHub Stars 100以上
0/15
○最近の活動
3ヶ月以内に更新がある
0/10
○フォーク
10回以上フォークされている
0/5
✓Issue管理
オープンIssueが50未満
+5
✓言語
プログラミング言語が設定されている
+5
○タグ
1つ以上のタグが設定されている
0/5
レビュー
💬
レビュー機能は近日公開予定です