What problem does it solve? Developers decide whether to try a project within seconds of landing on its README, yet most READMEs bury the install path, skip differentiation, and carry unverifiable claims. This Skill diagnoses the existing README against evidence, then rewrites it so a stranger can evaluate the project quickly and accurately. ## Core Features & Use Cases - Evidence-based audit: Runs a mechanical script plus a five-blocking-check, eight-scored-check scorecard, with every verdict backed by a line number, quoted sentence, or executed command. - Claim verification: Checks every command, flag, default, and behavioural assertion against the actual source code, and runs the documented install from a clean state. - Structured rewrite: Orders sections by cognitive funneling, calibrates badges using published research, fixes repository furniture, and adds enterprise trust signals only when the audience requires them. - Use Case: Before an open-source launch, ask for a README audit; receive a scorecard showing a broken install path and missing differentiation, then a section-by-section rewrite approved incrementally. ## Quick Start Ask the assistant to audit the README of your repository and produce the scorecard before rewriting anything.
