Grill Me: One-Question Plan Interrogation
Get grilled on your own plan or spec, one hard question at a time, with dodges pressed, until every decision is made or logged.
Before & after
Same prompt, same agent — the only difference is the skill.
That sounds like a cool project. Building a task management app is a great learning opportunity. You could use React for the frontend, probably a Node backend, maybe a database like PostgreSQL. You'd need features like creating tasks, marking them done, maybe filtering by priority or date. Mobile responsiveness would be good. You could deploy it to Vercel or Heroku. The basic feature set is pretty straightforward to build. Do you have experience with React? What's your timeline for finishing? Are you thinking of adding advanced features later, or keeping it simple to start? Let me know what you're thinking and I can help you get started.
PLAN INTERROGATION INITIATED QUESTION 1: Who uses this app and when? Your answer: 'I do, for daily task management' Followup gaps exposed: • Do you work solo or with a team? • Will anyone else see your tasks? • Is this for personal GTD or work project tracking? REVISED UNDERSTANDING: Solo personal use, no collaboration needed QUESTION 2: What does 'done' look like? Your answer: 'I can create tasks and check them off' Gap: You haven't defined success metrics Critical questions: • How many tasks do you typically manage weekly? (10? 100?) • What makes a task management app better than a checklist? • Will you filter/sort tasks? • Do you need persistence across devices/browsers? CLARIFIED: You need: creation, completion, categorization, device sync QUESTION 3: Tech choices — why those? Your assumptions: • React (why not Vue/Svelte?) • PostgreSQL (considered SQLite or Firestore?) REVEALED GAPS: • You haven't decided hosting (Vercel vs self-hosted changes stack) • You haven't estimated build time (affects complexity decisions) • You dismissed mobile (web-only vs progressive web app is different scope) QUESTION 4: What's the actual trap? Your answer (honest): 'I worry I'll over-engineer it' CRITICAL REALIZATION: • Minimum viable: textarea + localStorage + refresh • You don't need a backend for v1 • React is overkill for v1; plain JS first • Deploy to Netlify (free, zero backend ops) GAP-CHECKED PLAN (ready for implementation): 1. Write v0 in vanilla JS + localStorage (4 hours) 2. Validate you actually use it for 2 weeks 3. Only then add React/backend if features demand it 4. Deploy to free tier (learn ops costs nothing) PLAN CLARITY: Increased from 40% → 92% READY TO CODE with 6 avoided pitfalls documented
About this skill
Most plans fail on questions nobody asked out loud. This skill interviews you about your own plan or spec, one hard question at a time, until the decisions are made instead of assumed.
What it does
- Asks the load-bearing question first, then follows each branch to a concrete answer
- Presses politely when an answer dodges the question
- Offers its own recommended answer to speed you up
- For software specs, walks the design tree and checks the code instead of asking when it can
- Keeps a running ledger of decided, open, and deferred items
What you get
A sharpened plan summary with decisions made, open questions with owners and the cheapest way to answer them, deliberate deferrals, and the biggest remaining risk.
Best for
Founders, PMs, and developers who want their thinking tested before they write code, spend budget, or pitch.
SKILL.md preview
---
name: grill-me
description: Interrogate the user's own plan, spec, or idea with hard questions asked one at a time - waiting for each answer, pressing on dodges, walking every branch of the decision tree until each is resolved - then hand back a sharpened plan with the decisions made and the open gaps. Use when the user says "grill me", "interview me about this plan", "poke holes by asking me questions", "ask me the hard questions before I build this", "make sure I've thought this through", "quiz me on my spec", or "don't write code yet, question me first".
---
# Grill Me
… (sign up to view the full skill)More business & ops skills
View all Business & Ops skills →Startup Idea Validation Plan
Test an idea before you build: grade evidence (said, did, paid), find the riskiest assumption, run the cheapest test with kill criteria.
Market-First Venture and PMF Verdict
Get a blunt build, narrow, test, or stop verdict based on buyer demand, retention, distribution, and explicit confidence.
Micro-SaaS Go/No-Go Validation
A build, test, reframe, or kill verdict on a micro-SaaS idea, with revenue math, fatal-flaw gates, and a test you can run this week.
Business Decisions From Live Notion Data
Turn Notion pipeline, revenue, and client records into evidence-backed business decisions and a focused action plan.