Extract and rank risky assumptions from a spec. Loop until each has a validation method assigned.
Start the "Assumption Mapper" loop. - Goal: Every risky assumption in the spec has been identified, ranked by impact x uncertainty, and assigned a concrete validation method. - Max iterations: 5 - Between iterations run: `cat ASSUMPTIONS.md` - Exit when: ASSUMPTIONS.md has no rows with "validation: TBD" and at least 5 assumptions are documented with validation methods. Steps: 1. Read the spec/PRD document in this project 2. Extract implicit assumptions (about users, about technical constraints, about market conditions) 3. For each assumption, assess: impact if wrong (high/med/low) and certainty level (high/med/low) 4. Rank by risk = high-impact + low-certainty 5. For the top unvalidated assumptions, assign a validation method (user interview, A/B test, prototype test, data query, competitor analysis) 6. Write/update ASSUMPTIONS.md with the ranked table 7. If any assumption still has "validation: TBD", continue Self-pacing instructions: After each iteration, run the check command, read the output, and only continue if the exit condition is not met. Stop when exit condition passes or max iterations reached. Give a short status update each pass.
How it works