The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were identified in the skill instructions or reference materials.\n- [PROMPT_INJECTION]: The skill does not contain instructions that attempt to override agent behavior or bypass safety guidelines. Instructions focus on technical implementation details for the Astro framework.\n- [DATA_EXFILTRATION]: No evidence of hardcoded credentials, sensitive file path access, or unauthorized network operations was found. References to external URLs are limited to official documentation sites.\n- [OBFUSCATION]: The content is presented in clear markdown and code snippets without the use of Base64 encoding, zero-width characters, or homoglyph-based evasion techniques.\n- [COMMAND_EXECUTION]: The skill mentions standard development tools like astro check and astro sync within the context of framework workflows, but does not execute arbitrary or dangerous shell commands.\n- [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing untrusted user input and provides appropriate safety guidance.\n
Ingestion points: The input object in defineAction handlers.\n
Boundary markers: The skill mandates Zod schema validation at the server boundary.\n
Capability inventory: Handlers perform server-side mutations including database writes and sending emails.\n
Sanitization: Instructions recommend using parameterized queries, ORMs, and escaping user content before rendering.

astro-actions