The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides instructions for executing development-related shell commands including npx create-expo-module, npm install, and pod install. It also utilizes environment variables like EXPO_NONINTERACTIVE=1 to control CLI behavior.
[EXTERNAL_DOWNLOADS]: Scaffolding workflows trigger the download of project templates and library dependencies from the official NPM registry and GitHub repositories associated with the Expo organization.
[INDIRECT_PROMPT_INJECTION]: The skill exhibits an injection surface where user-provided data influences system-level operations.
Ingestion points: User-specified module slugs, paths, package names, and feature selections are intended to be incorporated into shell commands as documented in references/create-expo-module.md.
Boundary markers: The instructions do not define explicit delimiters or 'ignore' instructions to isolate user-provided strings from the command structure.
Capability inventory: The skill facilitates subprocess execution via npx and npm, as well as filesystem modifications through scaffolding tools.
Sanitization: There are no provided instructions for validating or escaping user-provided input before it is used in command-line arguments.
[METADATA_POISONING]: The SKILL.md frontmatter includes a risk: critical field. While this appears to be an author's self-assessment regarding the high-privilege nature of native module development (modifying manifests and native code), it is technically misleading as the skill's own logic and instructions do not present an inherent critical security risk beyond standard developer tooling behavior.

expo-module