goal-queue-workflow
Pass
Audited by Gen Agent Trust Hub on Aug 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill requires explicit human approval for all task definitions and budgets before they are enqueued, ensuring that the AI agent only operates within predefined constraints.\n- [EXTERNAL_DOWNLOADS]: Prerequisite guidance references the author's own repository (FuDesign2008/open-skills) for dependency management. This is a legitimate vendor-owned resource and is not used for unauthorized remote code execution.\n- [COMMAND_EXECUTION]: The skill utilizes dedicated Git branches for each task run to ensure isolation and prevent unintended side effects on the main codebase.\n- [SAFE]: Mandatory budget clauses and queue-level time caps are enforced to prevent runaway night runs or excessive resource consumption during unattended operation.
Audit Metadata