deep-interview

Convert vague AI missions into executable specs via Socratic questioning and ambiguity scoring.

3|Updated Apr 2, 2026
One-click install
npx skills add https://github.com/marcmunoz-uno/jailbreak --skill deep-interview-marcmunoz-uno
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: deep-interview
Source: https://github.com/marcmunoz-uno/jailbreak/tree/main/skills/deep-interview
Command: npx skills add https://github.com/marcmunoz-uno/jailbreak --skill deep-interview-marcmunoz-uno

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Deep Interview prevents premature autonomous execution by converting vague or shifting intents into crystal-clear specifications using Socratic questioning and mathematical ambiguity gating.

Core Features & Use Cases

  • Socratic deep requirements discovery: Asks one question at a time to expose hidden assumptions and stabilize the core goal, constraints, and success criteria.
  • Mathematical ambiguity scoring with explicit targeting: Scores clarity across dimensions, identifies the weakest dimension each round, and explains why the next question targets it.
  • Brownfield-aware context validation: For repo-changing missions, gathers codebase facts via an explore agent before confirming assumptions, reducing unsafe or misaligned modifications.
  • Pipeline-ready outputs: Produces a spec that is designed to feed directly into the next stages (consensus refinement and autonomous execution) only once ambiguity is sufficiently low.

Quick Start

Use the deep-interview skill to clarify your vague idea by running: "deep-interview --standard --autoresearch <your mission or vague idea>".

Frequently Asked Questions about deep-interview

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I turn vague ideas into executable specs before autonomous execution?▼

Executable specs are generated by running Socratic questioning on vague ideas, using dimension-based ambiguity scoring and gating to expose hidden assumptions and defer execution until clarity is achieved.

What is Socratic questioning for requirements discovery?▼

Socratic questioning for requirements discovery is a technique that asks one question at a time to stabilize core goals, constraints, and success criteria by exposing hidden assumptions before autonomous execution.

How do I prevent premature autonomous execution from unsafe or misaligned modifications?▼

Premature autonomous execution is prevented by mathematical ambiguity gating with explicit targeting, which scores clarity across dimensions and blocks pipelines until ambiguity drops to a configurable threshold.

Does requirements discovery work with brownfield exploration for existing codebases?▼

Requirements discovery works with brownfield exploration by gathering codebase facts via an explore agent before confirming assumptions, reducing unsafe or misaligned modifications to existing repositories.

How do I score ambiguity across dimensions to crystallize specifications?▼

Ambiguity is scored across dimensions by evaluating clarity, identifying the weakest dimension each round, and explaining why the next Socratic question targets it to crystallize specifications.