What problem does it solve? Teams often jump straight to building before properly framing the problem or weighing alternatives, leading to wasted effort on the wrong solution. This Skill runs a structured discovery interview that widens the option space, stress-tests candidate solutions, and captures the reasoning in a discovery document before any code is written. ## Core Features & Use Cases - Structured Discovery Interview: Asks one question at a time across framing, option generation, and second-order effect phases, withholding recommendations during framing and offering them during evaluation. - Calibrated Depth: Scales between a lightweight pass for single-feature choices and a heavyweight exploration for large feature sets or subsystems. - Discovery Summary Output: Writes a lean discovery.md document that feeds directly into the /spec step of the spec → plan → implement workflow. - Use Case: A team is unsure whether to add caching at the application or database layer. Run explore to frame the actual pain, generate both options plus alternatives, probe trade-offs and knock-on effects, and hand a decision-ready summary to /spec. ## Quick Start Ask the AI to run the explore skill on the problem or feature you want to investigate, then answer its questions one at a time.
