openclaw-debugging

Pass

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
  • [COMMAND_EXECUTION]: The instructions direct the agent to run project-specific commands like openclaw gateway and pnpm build to verify system behavior during debugging.\n- [DATA_EXFILTRATION]: The skill mentions checking for existing API keys in the user's ~/.profile file but explicitly commands the agent to never print secrets, mitigating the risk of credential exposure.
Audit Metadata
Risk Level
SAFE
Analyzed
May 14, 2026, 06:55 AM
Security Audit — agent-trust-hub — openclaw-debugging