The Agent Skills Directory

[SAFE]: The skill is composed entirely of markdown instructions. There are no executable scripts, binaries, or automated shell commands provided.
[SAFE]: No obfuscation techniques, hidden URLs, or suspicious encoding methods were detected.
[SAFE]: The skill does not perform any network operations, data exfiltration, or unauthorized file access. The mention of reading local project context files is a standard feature for AI agents to understand the specific product domain.
[SAFE]: While the skill discusses processing external data (Indirect Prompt Injection surface), it explicitly provides guidance on grounding AI outputs in structured data and using human/automated QA to ensure accuracy and safety.
[SAFE]: No privilege escalation or persistence mechanisms are included.

programmatic-seo