What problem does it solve? Verifying that an Expo skill actually works — that its description triggers it, that it generates compiling code, and that the resulting app renders correctly on real devices — requires a slow, manual loop of building fixtures, running apps, and taking screenshots. This Skill automates that entire evaluation pipeline end-to-end. ## Core Features & Use Cases - Trigger accuracy testing: Runs realistic prompts through claude -p subprocesses and measures recall of whether the skill under test gets triggered. - With/without-skill comparison: Generates code in fresh Expo fixtures both with and without the skill, then grades the delta with static gates (tsc, lint, expo export) and a design rubric. - Runtime screenshots: Boots iOS simulators and Android emulators, runs the generated app in Expo Go or a dev build, captures screenshots, and renders a self-contained HTML viewer comparing results. - Use Case: You edited the expo-ui skill's description and want to know if it still triggers and produces working SwiftUI screens. Run this Skill to get trigger recall, build pass/fail, side-by-side screenshots, and a graded quality report. ## Quick Start Ask the agent to evaluate the expo-ui skill end-to-end with runtime screenshots on iOS and Android using the latest Expo SDK.
