The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill provides instructions to download and install official tools like the Netlify CLI and various development dependencies from well-known registries like npm. These resources originate from trusted organizations and well-known services.
[COMMAND_EXECUTION]: Includes scripts such as check_deployment.sh, setup_env_vars.sh, and test_function_locally.sh that utilize the Netlify CLI to perform administrative tasks, including checking deployment status and setting environment variables.
[PROMPT_INJECTION]: The skill is designed to process untrusted external input from webhooks (e.g., SMS messages). This presents an indirect prompt injection surface where malicious data could influence downstream AI processing. The skill addresses this vulnerability by providing robust evidence and mitigation patterns: 1. Ingestion points: Webhook handlers in SKILL.md and assets/examples/webhook-function.ts read the event.body of incoming requests. 2. Boundary markers: The skill explicitly recommends and provides code for signature validation headers (e.g., telnyx-signature-ed25519). 3. Capability inventory: Functions utilize fetch for triggering background tasks and the Supabase client for database persistence. 4. Sanitization: Recommends the use of Zod for schema validation of incoming payloads in Quick Reference #10 to ensure data integrity.

netlify-integration