plans

Pass

Audited by Gen Agent Trust Hub on Apr 5, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No security risks identified. The skill implements a restrictive policy that limits the agent to read-only codebase exploration.
  • [SAFE]: Explicitly prohibits commands that could modify the system state, such as file creation, deletion, or package installation.
  • [SAFE]: Prevents data exfiltration and command chaining by forbidding the use of shell redirect operators (>, >>, |) and heredocs.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 5, 2026, 05:14 PM