The Agent Skills Directory

[SAFE]: The skill provides legitimate architectural guidance for Expo/EAS projects. The instructions are transparent and align with standard developer workflows.
[EXTERNAL_DOWNLOADS]: The skill references the execution of several official utility packages via npx, including create-expo-app, submit-expo-feedback, and the expo CLI. These are recognized as official vendor resources owned by 'expo'.
[COMMAND_EXECUTION]: Standard command-line operations for project initialization, package management (npx expo install), and feedback submission are documented. These are restricted to the intended purpose of the skill and use official tooling.
[INDIRECT_PROMPT_INJECTION]: The skill reads project configuration files (package.json, app.json) and user-supplied goals to determine routing.
Ingestion points: Reads package.json, app.json, and app.config.{js,ts}.
Boundary markers: None explicitly defined for file reads.
Capability inventory: Logic is limited to routing (suggesting other skills) and environment checks.
Sanitization: Standard validation of SDK versions is suggested. The risk is negligible as the data is used for classification rather than direct prompt interpolation in a high-privilege context.

expo-overview