interview
Aligns on a request's scope before planning — researches the codebase, surfaces unknowns, non-goals, and success criteria, asks batched clarifying questions only when research cannot resolve them, and writes a confirmed `brief.md`. Triggers on "align on scope", "interview me", "clarify the request", "/interview".
mthines · 已有静态检查记录
安装与使用
运行命令前先阅读来源与依赖说明。复制命令不会自动安装或执行。
npx skills add mthines/agent-skills --skill interview技能说明
name: interview
description: >
Aligns on a request's scope before planning — researches the codebase,
surfaces unknowns, non-goals, and success criteria, asks batched clarifying
questions only when research cannot resolve them, and writes a confirmed
brief.md. Triggers on "align on scope", "interview me", "clarify the
request", "/interview".
disable-model-invocation: false
argument-hint: '[] [--non-interactive]'
license: MIT
metadata:
author: mthines
version: '1.0.0'
workflow_type: advisory
tags:
- requirements
- scope-alignment
- elicitation
- clarifying-questions
- interview
- discovery
- pre-planning
- non-goals
- readiness-gate
- brief-artifact
Interview
Turns an underspecified request into a confirmed, thought-through scope before a single line is planned or written.
It restates and diffs the request, researches the code so its questions are specific, surfaces the unknowns and non-goals, interviews the user only where research falls short, and emits a brief.md that planning consumes.
This
SKILL.mdis a thin index. Phase procedures live inrules/*.mdand load on demand. A full worked run lives inreferences/interview-example.md.
Contents
When it runs
$ARGUMENTS is the request to align on (free text).
If $ARGUMENTS is empty, use the current conversation's active request; if there is none, ask what to scope — never interview against a guessed request.
| Flag | Default | Meaning |
|---|---|---|
--non-interactive | off | Do not prompt the user. Surface the questions inside brief.md under Open questions, proceed on stated assumptions, and set the readiness verdict. For headless callers that cannot prompt. |
This skill is adaptive.
When the request is already crisp and research answers the open questions, it asks nothing, writes a one-line pass note plus the brief, and returns ready.
It interrupts only for a genuine, load-bearing unknown — see rules/readiness-gate.md.
Relation to aw and ideate (precedence)
- Autonomous-implementation asks ("implement X autonomously", "ship this",
/aw) are owned byaw, which runs this interview inside its Phase 0 (on demand for Full tier — seerules/aw-integration.md) — do not also fire standalone; letawdrive. - Standalone
/interviewis for the explicit "help me scope this first" moment on non-awor ambiguous work. - For solution options, not scope, use
/ideate— it diverges and scores; this skill converges on the one scope.
Workflow
Five phases, each with a gate. Do not advance until the gate passes.
| Phase | Name | Rule file | Gate |
|---|---|---|---|
| 0 | Restate & diff | rules/research-grounding.md | Request restated in own words; every delta from the user's words surfaced. |
| 1 | Research the ground | rules/research-grounding.md | Affected areas, prior art, patterns, and constraints located in the code. |
| 2 | Surface unknowns (gate) | rules/readiness-gate.md, rules/consultative-completeness.md | Every unknown enumerated and classified blocking vs advisory; interview-or-pass decided. |
| 3 | Interview | rules/question-design.md | Blocking + high-value unknowns asked in ≤ 2 batched rounds, or skipped (adaptive pass). |
| 4 | Confirm & brief | rules/brief-artifact.md | Understanding confirmed; brief.md written; readiness verdict set. |
Required reading by phase
Load on demand — do not preload.
| Phase | Files |
|---|---|
| 0–1 | rules/research-grounding.md |
| 2 | rules/readiness-gate.md, rules/consultative-completeness.md |
| 3 | rules/question-design.md |
| 4 | rules/brief-artifact.md |
| aw | rules/aw-integration.md — how aw delegates to this skill. |
Core principles
- Research before asking. A question the codebase already answers is noise. Ground every question in what the code shows so it is specific ("reuse
PrStatusCacheor add a cache?"), not generic ("what do you want?"). Seerules/research-grounding.md. - **Interrupt