trellis-meta

Pass

Audited by Gen Agent Trust Hub on May 7, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is a meta-documentation tool designed to help AI agents understand and navigate the Trellis project structure.
  • [SAFE]: Instructions focus on local project customization within the .trellis/ and platform-specific directories (e.g., .claude/, .cursor/).
  • [SAFE]: The skill explicitly advises against modifying global system directories or upstream source code unless requested by the user.
  • [SAFE]: No evidence of credential theft, hidden communication, or malicious persistence mechanisms was found.
  • [SAFE]: The documented hooks and scripts (e.g., session-start.py, ralph-loop.py) are part of the Trellis system's standard lifecycle management and context injection mechanism as described in the architectural references.
Audit Metadata
Risk Level
SAFE
Analyzed
May 7, 2026, 06:20 PM