The Agent Skills Directory

[SAFE]: Comprehensive analysis of the instructions, evaluation cases, and reference files confirms the skill is entirely instructional and contains no malicious patterns or prompt injections.- [SAFE]: The skill does not perform any network operations, command executions, or sensitive data access.- [SAFE]: The skill's reference to a local context file (.agents/product-marketing-context.md) for personalization is a benign and standard operational pattern.- [SAFE]: No obfuscation, persistence mechanisms, or privilege escalation vectors were identified in any of the provided files.

programmatic-seo