home / skills / yellinzero / aico / aico-pm-clarification

aico-pm-clarification skill

/.codex/skills/aico-pm-clarification

This skill guides requirement clarification by asking one focused question at a time, offering recommended options and reasoning to prevent overload.

npx playbooks add skill yellinzero/aico --skill aico-pm-clarification

Review the files below or copy the command above to add this skill to your agents.

Files (1)
SKILL.md
2.6 KB
---
name: aico-pm-clarification
description: |
  Resolve requirement ambiguities through STRUCTURED questioning: one question at a time, with recommended options and reasoning.

  UNIQUE VALUE: Prevents overwhelming users with multiple questions. Provides expert recommendations for each decision.

  Use this skill when:
  - Running /pm.clarify command
  - User says "unclear", "not sure what this means", "confused about"
  - User asks "what does X mean?", "how should X work?", "can you clarify?"
  - Requirements have conflicting or inconsistent details
  - Stories are missing acceptance criteria or have gaps
  - Need to fill information gaps BEFORE development can proceed

  Process: Ask ONE question at a time (max 5 per session), provide recommended option with reasoning.
  DO NOT ask multiple questions at once - this overwhelms users.
---

# Requirement Clarification

## Language Configuration

Before generating any content, check `aico.json` in project root for `language` field to determine the output language. If not set, default to English.

## Process

1. **Scan context**: Check `docs/reference/pm/` for existing documentation
2. **Identify ambiguities**: Categorize by type (scope, behavior, data, edge cases)
3. **Prioritize**: Sort by impact: scope > security > UX > technical
4. **Ask ONE question at a time**: Max 5 questions per session
5. **Provide recommendation**: Each question should have a recommended option with reasoning
6. **Update docs**: Document each answer immediately

## Question Format

```markdown
### Question [N]: [Topic]

**Context**: [Quote relevant requirement]

**Ambiguity**: [What's unclear]

**Options**:
| Option | Description | Implications |
|--------|-------------|--------------|
| A | [First option] | [Trade-offs] |
| B | [Second option] | [Trade-offs] |

**Recommended**: [Option] because [reasoning]
```

## Ambiguity Categories

| Category       | Focus                      |
| -------------- | -------------------------- |
| Scope          | What's included/excluded   |
| Behavior       | How feature should work    |
| Data           | What information is needed |
| Edge cases     | Unusual scenarios          |
| Error handling | Failure modes              |

## Key Rules

- ALWAYS ask ONE question per message, never batch
- MUST provide recommended option with reasoning for each question
- ALWAYS prioritize blocking issues (scope, security) over minor details
- Max 5 questions per clarification session

## Common Mistakes

- ❌ Ask all questions at once → ✅ One at a time
- ❌ Open-ended questions → ✅ Multiple choice with recommendation
- ❌ Low-impact questions → ✅ Focus on blocking issues first

Overview

This skill resolves requirement ambiguities through a structured, one-question-at-a-time approach. It prevents overwhelming stakeholders by asking a single focused question per message, offering recommended options and the reasoning behind each choice. The goal is to surface blocking decisions so development can proceed with clear acceptance criteria.

How this skill works

The skill scans project context and requirement docs to identify ambiguous items and categorizes them by scope, behavior, data, edge cases, or error handling. For each ambiguity it poses exactly one multiple-choice question with a recommended option and concise reasoning. It limits sessions to a maximum of five questions and records answers to update documentation immediately.

When to use it

  • When a user runs /pm.clarify or requests clarification in chat
  • If a requirement contains conflicting or missing details
  • When stories lack acceptance criteria or have unclear behavior
  • If stakeholders say 'unclear', 'not sure what this means', or 'confused about'
  • Before development begins and information gaps block progress

Best practices

  • Ask only one focused question per message to avoid cognitive overload
  • Prioritize blocking issues first: scope and security, then UX and technical details
  • Provide 2–3 clear options for each question and state trade-offs
  • Always include a recommended option and short reasoning to guide decision-making
  • Document each resolved ambiguity immediately in project docs

Example use cases

  • Clarifying whether a feature includes admin-only controls or is available to all users
  • Deciding expected behavior for ambiguous edge cases (e.g., empty inputs or timeouts)
  • Choosing between data retention durations when requirements omit storage rules
  • Resolving conflicting acceptance criteria between product and engineering
  • Filling gaps in stories so QA can write deterministic tests

FAQ

How many questions will you ask in one session?

Up to five questions per clarification session, with only one question asked at a time.

What format do questions use?

Each question is multiple choice with a short context quote, ambiguity description, options with implications, and a recommended option with reasoning.