skills/pr-pm/prpm/human-writing/Gen Agent Trust Hub

human-writing

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill's primary content consists of writing guidelines, tone calibration, and structural techniques for creating human-like content. No malicious instructions, prompt injections, or data exfiltration patterns were detected.
  • [COMMAND_EXECUTION]: The documentation includes example shell commands for the 'prpm' tool (e.g., prpm install, prpm init). These commands are identified as vendor resources belonging to the author 'pr-pm' and are used strictly within a tutorial context.
  • [EXTERNAL_DOWNLOADS]: The skill references official project links, including a GitHub repository for issue tracking and a Twitter account for community engagement, which are legitimate project resources.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 06:17 PM