EXECUTE IMMEDIATELY

Step 0: Check for Prior Brainstorm Output (Advisory)

Look for brainstorm output in the current session:

If .shinchan-docs/*/brainstorm-output.md exists: Read the brainstorm output. Note to user: "Prior brainstorm found — using it to seed the requirements interview." Include in Misae's prompt: "A brainstorm was completed. Summary: [brainstorm-output content]. Use this as context for the requirements interview." Else: Note to user: "No prior brainstorm found. For deeper requirements coverage, consider running /team-shinchan:brainstorm first. Continuing with requirements interview." (Continue to Step 1 regardless — brainstorm is optional)

Step 1: Validate Input

If args is empty or only whitespace:
  Ask user: "What feature or task would you like me to analyze for hidden requirements?"
  STOP and wait for user response

If args length > 2000 characters:
  Truncate to 2000 characters
  Warn user: "Request was truncated to 2000 characters"

Step 2: Execute Task

Do not read further. Execute this Task NOW:

Task(
  subagent_type="team-shinchan:misae",
  model="sonnet",
  prompt=`/team-shinchan:requirements has been invoked.

## Hidden Requirements Analysis Request

Analyze the request to discover:

| Category | Analysis Focus |
|----------|----------------|
| Hidden Requirements | Unstated needs, implicit expectations |
| Edge Cases | Boundary conditions, unusual inputs |
| Risks | Potential problems, failure scenarios |
| Dependencies | Prerequisites, blocking items |
| Scope | Clarifications needed, ambiguities |

## Analysis Areas

- Error handling scenarios
- Performance implications
- Security considerations
- User experience impacts
- Maintenance burden
- Integration points

## Output Requirements

- Show analysis process in real-time
- List all hidden requirements found
- Prioritize risks by impact
- Provide specific recommendations
- Note any clarifications needed from user

User request: ${args || '(Please describe what to analyze)'}
`
)

STOP HERE. The above Task handles everything.