pragman-shapelisted

# Pragman Shape Turn an ambiguous opportunity into the smallest valuable, testable contract. Preserve uncertainty where evidence is weak and prevent solution detail from outrunning the problem. ## Use this skill when - A request starts as a feature idea, roadmap item, bug, or ambition but its outcome and boundary are unclear. - You need to decide what not to build, which assumption matters most, and what evidence would kill the bet. - Implementation is tempting before the user, constraints, dependencies, or approval boundary are understood. ## Acceleration payoff It converts ambiguity into a small learning loop so teams stop building roadmap-sized guesses and start with evidence-producing work. ## Workflow 1. Name the problem, intended user, triggering situation, and why it matters now. Separate observed evidence from assumptions and requested solutions. 2. Ask focused questions only where the answer can change the desired outcome, smallest bet, kill criterion, safety boundary, or provider route. Use a configured brainstorming or product specialist through `pragman-router` when it adds leverage. 3. State the desired outcome as a change in user or system behavior, not a feature list. Identify explicit non-goals and the current baseline. 4. Choose the smallest valuable experiment or deliverable that can resolve the highest-risk assumption. Prefer a reversible slice with a short learning loop over a broad foundation. 5. Define success evidence, failure evidence, and a kil