expo-skill-eval
Pass
Audited by Gen Agent Trust Hub on Jun 24, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a robust evaluation pipeline that manages temporary workspaces in
/tmp/expo-skill-eval-*and cleanups up multi-GB build artifacts (DerivedData, Android Gradle output) after each test run to maintain system stability. - [COMMAND_EXECUTION]: Uses standard development tools including
bun,expo,adb,xcrun simctl, andgitto automate the build and testing of generated React Native applications. - [EXTERNAL_DOWNLOADS]: Fetches official Expo templates and project dependencies from the npm registry. It also downloads official browser binaries via Playwright for web-based screenshots.
- [DATA_EXFILTRATION]: Includes a feature to publish results to claude.ai Artifacts for sharing. This feature is opt-in and requires explicit user confirmation before any data is sent externally.
- [REMOTE_CODE_EXECUTION]: Executes code generated by sub-agents within isolated fixture environments. This behavior is the central purpose of the skill and is documented as a mechanism to verify code quality and runtime correctness.
Audit Metadata