---
name: wicked-garden-product-requirements-analysis
description: |
  Use when turning a vague idea or stakeholder ask into structured user stories with acceptance criteria
  and a requirements graph. For complexity >= 3 or compliance projects, defaults to graph mode.
  NOT for navigating an existing requirements graph (use requirements-navigate) or stakeholder alignment (use product:align).
phase_relevance: ["clarify", "design", "review"]
archetype_relevance: ["*"]
---

# Requirements Analysis Skill

Structured requirements elicitation from discovery through documentation.

> **Default: Graph mode.** For complexity >= 3 or compliance projects, produce
> a `requirements/` graph (see `requirements-graph` skill) instead of a
> monolithic document. Use `--monolith` to force the legacy single-file format.
> See: `requirements-graph/SKILL.md` for graph layout and schema.
> See: `requirements-navigate/SKILL.md` for querying and maintaining the graph.

## Process

### 1. Context Gathering
- Read project brief, outcome.md, or requirements
- Identify existing documentation
- Note assumptions and constraints

### 2. Stakeholder Identification
- Who are the users/personas?
- Who are secondary stakeholders?
- Who has decision authority?

### 3. Requirements Elicitation

Ask the 5 W's:
- **Who**: Which users/personas?
- **What**: What capabilities needed?
- **When**: What triggers the need?
- **Where**: What context/environment?
- **Why**: What value/benefit?

### 4. User Story Definition

Format: **As a [persona], I want [capability], so that [benefit]**

Quality criteria:
- Specific persona identified
- Capability clearly stated
- Benefit/value articulated
- Testable outcome

### 5. Acceptance Criteria

For each story:
```
Given [context/precondition]
When [action/event]
Then [expected outcome]
```

Include:
- Happy path scenarios
- Error conditions
- Edge cases
- Non-functional requirements

### 6. Validation

Check:
- [ ] All stories have persona, capability, benefit
- [ ] Acceptance criteria are testable
- [ ] Edge cases identified
- [ ] Dependencies mapped
- [ ] Ambiguities flagged

## User Story Template

```markdown
### US{N}: {Story Title}

**As a** {specific persona}
**I want** {specific capability}
**So that** {specific benefit}

**Priority**: {P0/P1/P2}
**Complexity**: {S/M/L/XL}

**Acceptance Criteria**:
1. Given {context}, When {action}, Then {outcome}
2. Given {error condition}, When {action}, Then {error handling}
3. Given {edge case}, When {action}, Then {graceful handling}

**Non-Functional**:
- Performance: {requirement}
- Security: {requirement}
- Usability: {requirement}

**Dependencies**: {Other stories or systems}
**Assumptions**: {What we're assuming}
**Open Questions**: {What needs clarification}
```

## Common Patterns

**CRUD Operations**:
- Create: As a user, I want to create X, so that I can track Y
- Read: As a user, I want to view X, so that I can understand Y
- Update: As a user, I want to edit X, so that I can correct Y
- Delete: As a user, I want to remove X, so that I can clean up Y

**Authentication/Authorization**:
- Login, logout, password reset
- Role-based access control
- Session management

**Data Validation**:
- Input validation
- Error messaging
- Constraint enforcement

## Integration with Tools

Dispatch uses the Skill tool on Claude Code (a fresh forked context). On any other harness, open the named skill's `SKILL.md` from your skills catalog and carry out its instructions inline with the given args, then continue here.

```bash
# Search for similar requirements (FTS5 over indexed docs + wiki)
Skill(skill="wicked-garden-mem", args="recall \"user story\"")

# Store requirements on the active clarify task via native TaskUpdate
# TaskUpdate(taskId="{task_id}", description="{previous}\n\n## Requirements\n{user_stories}")

# Recall past patterns
Skill(skill="wicked-garden-mem", args="recall \"requirements for {feature_type}\"")
```

## Output Format

See: `refs/requirements-output-format-template.md`, `refs/requirements-example-auth.md`, and `refs/requirements-example-export-integration.md`