The Agent Skills Directory

[SAFE]: The skill does not contain any malicious code, prompt injections, or obfuscation techniques. Its content is purely instructional documentation and code samples for developer guidance.
[SAFE]: The provided code examples demonstrate secure implementation patterns. Specifically, it uses Zod schemas to strictly validate input parameters (include and fields), preventing unauthorized access to unmodeled fields.
[SAFE]: The guidelines explicitly mandate permission checks (authorization) within the query handler before sensitive fields are loaded or returned to the client, which is a key defense against data exposure.
[SAFE]: The skill implements a 'least privilege' approach by restricting data expansion to a single level and encouraging a three-step flow (base query, authorized extension, field-level filtering) that minimizes the attack surface.

implement-trpc-query