The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill documents how to handle external inputs via HTTP requests, path parameters, and form events (e.g., in netlify/functions/travel.mts and netlify/functions/on-deploy.mts).
Ingestion points: Untrusted data enters the agent context through the Request object, context.params, and event.data.
Boundary markers: No explicit instructions are provided for using delimiters or warnings to isolate untrusted content.
Capability inventory: The functions demonstrate capabilities including network requests via fetch, database queries using @netlify/database, and access to environment variables via Netlify.env.get.
Sanitization: The provided examples do not demonstrate input validation or sanitization before processing external content.
[EXTERNAL_DOWNLOADS]: The skill documentation recommends the installation of official libraries such as @netlify/functions and @netlify/vite-plugin from a well-known service provider via NPM.
[COMMAND_EXECUTION]: Instructions describe the use of standard development tools like the Netlify CLI for local testing (netlify dev) and function invocation (netlify functions:invoke).

netlify-functions