| name | deep-interview |
| description | Run a structured, ambiguity-gated interview that turns an underspecified request into a decision-complete brief. |
Deep Interview
Use this skill when implementation would otherwise require material guesses about outcome, audience, constraints, workflow, or acceptance criteria.
Inputs
- The user request and available source material.
- The decision or artifact being prepared.
- Known constraints, stakeholders, and deadlines.
Protocol
- Inspect available evidence before asking questions.
- State the current interpretation in plain language.
- Ask one high-impact question at a time.
- Prefer concrete choices when they represent real tradeoffs; accept free-form answers when the space is genuinely open.
- Resolve intent first: goal, success, audience, boundaries, and constraints.
- Resolve implementation second: interfaces, data flow, failure modes, validation, rollout, and compatibility.
- Maintain a compact decision ledger rather than replaying the full interview.
- Stop when the brief is decision complete and no implementer choices remain.
Ambiguity Gate
Ask only when the answer changes scope, behavior, risk, or acceptance. Discoverable repository facts must be inspected, not asked. If a low-risk preference is unanswered, use the recommended default and record it as an assumption.
Output
Return a concise brief containing:
- Goal and success criteria.
- Audience and scope boundaries.
- Required behavior and interfaces.
- Source and privacy requirements.
- Edge cases and recovery behavior.
- Test and acceptance criteria.
- Explicit assumptions.
Supporting Material
Read the complete interview playbook only when a specialized interview mode, extended example, or full question bank is needed. Use workshop-facilitation for pacing.
