The Agent Skills Directory

[SAFE]: The skill is entirely documentation-based, providing technical guidance for the Zod validation library without any executable code or scripts that perform system operations.\n- [SAFE]: No prompt injection or behavior override attempts were found. The instructions are focused on ensuring the agent uses the correct Zod v4 API patterns instead of outdated v3 ones.\n- [SAFE]: Data security is emphasized within the skill. For example, the rules/error-input-security.md and rules/observe-structured-errors.md files explicitly warn users against enabling features that could leak sensitive data (PII) into logs or API responses.\n- [SAFE]: All external references are to official documentation or well-known developer tools (e.g., zod.dev, standard-schema, ESLint, GitHub Actions). No suspicious remote downloads or execution patterns were detected.\n- [SAFE]: The skill does not access sensitive local files or directories. It only provides examples of how a developer might use Zod to validate environment variables or form data within their own applications.

zod