doc-coauthoring

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements a multi-stage workflow designed for collaborative documentation authoring. It utilizes standard tool capabilities like create_file and str_replace to manage document drafts and iterative refinement, which align with its primary stated purpose.
  • [SAFE]: The skill incorporates security best practices by explicitly asking for user confirmation before accessing external tools, searching connected platforms, or reading content from provided links.
  • [SAFE]: No malicious patterns, obfuscation, or unauthorized data exfiltration were detected. The instructions for 'Reader Testing' involve utilizing platform features (sub-agents) or manual user testing in fresh sessions, effectively sandboxing context to ensure document clarity.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 02:39 PM
Security Audit — agent-trust-hub — doc-coauthoring