safety-check

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill allows the agent to run a local script '~/repos/github/my-repos/Claude-MD-Tester/bin/status.sh'. This is documented as a read-only pre-flight check to verify the system state.
  • [SAFE]: The instructions explicitly forbid the agent from executing the configuration swap scripts ('activate.sh' and 'restore.sh'), emphasizing that these destructive or recovery actions must be performed by the user in a separate terminal to ensure safety and maintain user control.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 07:17 PM
Security Audit — agent-trust-hub — safety-check