skillforge

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill includes several Python scripts (discover_skills.py, triage_skill_request.py, validate-skill.py) that handle local skill management. These scripts use standard libraries to scan directories and parse markdown files. The access is limited to agent-specific paths like .claude/skills and does not involve sensitive system files or credentials.
  • [SAFE]: No remote code execution patterns or suspicious network activities were detected. The triage system processes user input using regex and scoring for local routing decisions rather than dynamically executing external payloads.
  • [SAFE]: The methodology documentation and templates provided are benign and follow best practices for skill development and future-proofing.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 07:00 AM