The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to run a local utility script, scripts/list-components.js, to inspect the available components and modifiers in the project's @expo/ui installation. This is a standard developer convenience to provide context to the AI agent.\n- [EXTERNAL_DOWNLOADS]: The skill recommends installing the @expo/ui package via npx expo install, which is the standard procedure for Expo projects. It also references official documentation from docs.expo.dev and Material Design guidelines, which are well-known and trusted sources.\n- [INDIRECT_PROMPT_INJECTION]: The list-components.js script reads from the local node_modules/@expo/ui directory to generate its output, creating a potential surface for indirect prompt injection if the installed package were compromised. However, the script only extracts structured component metadata.\n
Ingestion points: scripts/list-components.js reads package.json and .d.ts files from the project's node_modules/@expo/ui folder.\n
Boundary markers: The script output does not include explicit delimiters or instructions to ignore embedded content.\n
Capability inventory: The script extracts component and modifier names and JSDoc summaries to provide context to the agent.\n
Sanitization: The script uses regex to filter for specific TypeScript export patterns and JSDoc blocks, limiting what content is ingested.

expo-ui