execute-fleet
Pass
Audited by Gen Agent Trust Hub on Aug 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill defines a robust orchestration workflow with multiple security 'Hard Gates', such as capability detection and git status verification before execution.
- [SAFE]: Interaction with external agents (workers) is restricted via a strict protocol. The orchestrator validates worker messages against a terminal status allowlist (DONE, BLOCKED, etc.) and verifies the presence and content of 'CDD cards' before accepting task completion.
- [SAFE]: Resource access is limited to project-specific paths like
.taskmaster/and.atlas-ai/. The skill does not attempt to access sensitive system files or environment credentials. - [SAFE]: The use of vendor-specific tools (mcp__atlas-*) and local scripts (script.py, ship-check.py) is consistent with the skill's stated purpose of managing a fleet pipeline.
Audit Metadata