What problem does it solve? After a round of customer interviews, the learning scatters across hypothesis files, question lists, and per-interview debriefs that nobody rereads. This Skill consolidates all of it into a single FINAL-REPORT.md that states what is now known, what is merely believed, and what remains untested, with every claim backed by tallies, debrief citations, and verbatim customer quotes. ## Core Features & Use Cases - Evidence-graded findings: Every finding is numbered (F1, F2, ...) and tagged validated, disproved, directional, watch, or untested, with exact voice tallies like "7 of 9" instead of vague probability words. - Two-layer report structure: A citation-free summary for skimmers on top, and a fully cited body with quotes and debrief references below for deep readers and downstream LLM work. - Per-area briefs: Marshals findings for ideal-customer definition, positioning and messaging, pricing and packaging, marketing and sales, and product priorities, without doing that downstream work itself. - Use Case: After completing eight customer discovery interviews with debriefs on file, ask for a report and receive a FINAL-REPORT.md the whole company can use, including sampling-bias disclosure and a vocabulary bank of the customers' own words. ## Quick Start Write up what we found from the interviews in this directory into a final report for the team.
