The Agent Skills Directory

[SAFE]: The skill provides standard documentation and code templates for the Astro Actions API, a common framework feature.
[SAFE]: Implementation patterns strongly emphasize security best practices, such as mandatory input validation using Zod schemas to prevent processing of malformed or malicious data.
[SAFE]: Guidance on authentication and error handling follows established secure development practices, such as verifying session cookies before executing server-side logic.
[INDIRECT_PROMPT_INJECTION]: The skill manages the attack surface associated with processing external form data by recommending robust validation boundaries.
Ingestion points: The skill defines handlers for FormData and JSON payloads in src/actions/index.ts as seen in multiple reference files.
Boundary markers: The recommended use of input: z.object({...}) provides a strict validation boundary for all ingested data.
Capability inventory: Handlers in the templates are shown performing typical server-side tasks such as database mutations and logging.
Sanitization: All input is parsed, typed, and validated by Zod prior to processing in the handler logic, effectively sanitizing external content.

astro-actions