The Agent Skills Directory

[COMMAND_EXECUTION]: The scripts/check-versions.sh script utilizes the npm view command to fetch metadata from the npm registry to verify if local package versions are current. This is a standard utility for developer environment maintenance.\n- [EXTERNAL_DOWNLOADS]: The skill documentation and templates reference several well-known packages from the official npm registry, including hono, zod, and valibot. These are established industry-standard tools from a trusted ecosystem.\n- [PROMPT_INJECTION]: The skill provides patterns for handling untrusted user input through Hono's request parsing methods (e.g., c.req.json(), c.req.query()), which constitutes an indirect prompt injection surface.\n
Ingestion points: Untrusted data enters the execution context through request parsing in SKILL.md and templates such as routing-patterns.ts.\n
Boundary markers: The skill explicitly instructs developers to use validation middleware like zValidator or vValidator and the c.req.valid() method to enforce strict data schemas.\n
Capability inventory: Code templates demonstrate network operations via fetch, logging, and simulated database interactions.\n
Sanitization: Validation best practices are the core focus, utilizing Zod/Valibot schemas to ensure all external data is sanitized and typed before use.

hono-routing