What problem does it solve? Verifying that an Expo skill actually works—triggers on the right prompts, generates compiling code, and renders correctly on real devices—requires a slow, manual testing loop. This Skill automates that end-to-end evaluation pipeline. ## Core Features & Use Cases - Trigger Accuracy Testing: Runs realistic prompts through claude -p subprocesses and measures whether the skill under test is selected, reporting recall. - Code Quality Gates: Generates apps with and without the skill in cached Expo fixtures, then runs tsc --noEmit, diff-aware lint, and expo export as a static gate. - Runtime Screenshots: Boots iOS simulators and Android emulators (or web via Playwright), runs the generated app in Expo Go or a development build, and captures screenshots for visual grading against expectations or a reference image. - Results Viewer: Generates a self-contained HTML report comparing with-skill vs without-skill scores, screenshots, and design-quality rubric dimensions. - Use Case: After editing a skill's description in plugins/expo/skills/, run an eval to confirm the new description improves trigger recall and that generated code still renders correctly on both platforms. ## Quick Start Evaluate the expo-ui skill end-to-end with runtime screenshots on iOS and Android using the latest Expo SDK.
