run-task
Pass
Audited by Gen Agent Trust Hub on Jun 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a bootstrap verification process to ensure it only operates within intended project environments.
- [SAFE]: Universal principles are integrated into the instructions to explicitly forbid agents from accessing or exposing credentials, API keys, or PII.
- [SAFE]: Data ingestion is limited to reading project manifests for environment detection, with no evidence of remote exfiltration or command injection.
- [SAFE]: Orchestration uses platform-standard tools for team management and communication, maintaining a controlled execution flow.
Audit Metadata