What problem does it solve? Plans often contain hidden undecided branches that surface mid-implementation as costly rework. This Skill walks a plan's decision tree before any code is written, forcing every open decision to be resolved, inferred from the codebase, or explicitly deferred. ## Core Features & Use Cases - Decision Tree Extraction: Converts a plan into nodes marked open or resolved, with each resolved node tagged by source (user answer or inferred from the codebase). - One-Question-at-a-Time Interrogation: Asks a single tight question per ready node, always paired with a recommended answer and one sentence of reasoning. - Codebase-First Resolution: Checks the repository, prior commits, and existing docs before asking, marking answers as inferred with a source path. - Use Case: Before implementing a new caching layer, run the interrogation to pin down storage backend, TTL policy, and invalidation strategy, producing a decision ledger to paste into the plan doc. ## Quick Start Ask the AI to stress-test your implementation plan using plan-interrogate before writing any code.