The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to guide an agent in building content-driven applications, which inherently involves processing untrusted external data.
Ingestion points: The skill utilizes the Astro Content Layer API, which ingests data from local files (Markdown, MDX, JSON, YAML) and remote sources via custom loaders or Live Loaders (as described in src/content.config.ts). It also handles user input via Astro.request and Astro.session in SSR modes.
Boundary markers: The instructions do not explicitly mandate the use of delimiters or specific "ignore embedded instructions" headers when rendering processed content, though the framework's architecture separates template logic from content.
Capability inventory: The skill empowers the agent to write filesystem components, configuration files, and server actions, and to perform network requests within data loaders.
Sanitization: The skill mitigates this risk by strictly requiring the use of Zod schemas (astro/zod) for all content collections to perform type validation and coercion, and by promoting the use of astro:env for secure environment variable management.
[COMMAND_EXECUTION]: The documentation includes standard Astro CLI commands used for project setup and maintenance.
Evidence: The skill references commands such as npx astro add for installing official adapters (e.g., Node, Vercel, Netlify, Cloudflare) and UI integrations (React, Vue, Svelte, etc.) in files like references/ssr-adapters.md and references/styling.md.

astro-framework