The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill provides instructions for submitting feedback using npx --yes submit-expo-feedback@latest, which retrieves content from the NPM registry.
[REMOTE_CODE_EXECUTION]: The feedback mechanism utilizes npx to download and execute the submit-expo-feedback package locally.
[COMMAND_EXECUTION]: The documentation includes a shell command for interacting with feedback services related to the Expo framework.
[PROMPT_INJECTION]: The framework described for DOM components presents a surface for indirect prompt injection where data is passed from native props into a browser context with access to native actions.
Ingestion points: Data props such as data, code, and content passed to DOM components as shown in SKILL.md.
Boundary markers: Absent; the documentation does not suggest delimiters or explicit instructions for the agent to ignore instructions embedded in props.
Capability inventory: DOM components are provided with capabilities to invoke native functions such as saveData (potential storage write) and showAlert, as defined in SKILL.md.
Sanitization: Absent; there is no guidance on sanitizing or validating input props before they are rendered or processed within the webview context.

expo-dom