cleanup-agent-team-node-processes
Pass
Audited by Gen Agent Trust Hub on Aug 12, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The script uses the Stop-Process cmdlet to terminate lingering agent-related processes. It mitigates risks by requiring specific inclusion patterns and verifying the process identity (matching name and creation time) before execution to prevent race conditions or PID reuse issues.
- [DATA_EXFILTRATION]: The skill reads process command lines which could potentially contain secrets. It incorporates a redaction function that masks common credential parameters such as tokens and API keys to prevent data exposure in audit logs or temporary files.
- [SAFE]: The skill enforces secure practices for file handling by requiring output paths to be within the operating system's temporary directory and ensuring these files are deleted in a finally block to prevent data lingering in the workspace.
Audit Metadata