JPeetz/agent-skills / skills/skill-development/interview-me
interview-me
VerifiedUse this skill when an ask is underspecified — missing who it's for, why now, or what success looks like. One-question-at-a-time interview to extract what the user actually wants (not what they think they should want) before any plan, spec, or code exists. Requires a live, responsive user — do not invoke in non-interactive contexts (CI runs, scheduled tasks, autonomous loops).
| File | Kind | Size |
|---|---|---|
| CHANGELOG.md | markdown | 458 B |
| LICENSE | other | 1.0 KB |
| SKILL.md | skill md | 10.7 KB |
| evals/evals.json | config | 2.2 KB |
evals/evals.json
View on GitHub[
{
"id": "ev001",
"description": "Hypothesis formulation — agent states a hypothesis with confidence number for an underspecified request",
"scenario": "User says 'build me a dashboard for our metrics'. Agent should respond with a hypothesis (one sentence) and a confidence number below 100% that includes what's missing.",
"expected_outcome": "Agent states HYPOTHESIS + CONFIDENCE ~30-40% with explicit mention of missing context (who it's for, what metrics, success criteria)"
},
{
"id": "ev002",
"description": "Single-question discipline — agent asks one question at a time with guess attached",
"scenario": "Following ev001, user responds to the hypothesis. Agent should ask ONE follow-up question with its GUESS attached, not batch multiple questions.",
"expected_outcome": "Agent asks exactly one question with GUESS: prefix, waits for response before next question"
},
{
"id": "ev003",
"description": "Want-vs-should-want detection — agent probes when user gives a sophistication-signaling answer",
"scenario": "User responds with 'I want it to be scalable and use clean architecture'. Agent should detect this as a sophistication-signaling answer and probe with 'If you didn't have to justify this to anyone, what would you actually want?'",
"expected_outcome": "Agent identifies the buzzword answer and deploys the 'justify to anyone' probe"
},
{
"id": "ev004",
"description": "Structured restate — agent writes back a concrete restate with Out of scope",
"scenario": "After several rounds of Q&A, user's intent becomes clear. Agent should restate the intent using Outcome/User/Why now/Success/Constraint/Out of scope format.",
"expected_outcome": "Agent produces a structured restate with all 6 fields, including Out of scope, and asks for explicit confirmation"
},
{
"id": "ev005",
"description": "Terminal stop — agent does NOT start work after receiving explicit confirmation",
"scenario": "User confirms the restate with an explicit 'yes'. Agent should stop its turn without invoking any tools or starting downstream work.",
"expected_outcome": "Agent delivers confirmed statement of intent, offers downstream paths, and stops turn — no tool invocations follow"
}
]Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |