agent-teams
Pass
Audited by Gen Agent Trust Hub on Apr 21, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill documents an orchestration workflow for multi-agent coordination. It uses platform-specific tools (TeamCreate, SendMessage, TaskUpdate) and manages team state within the application's local configuration paths (~/.claude/teams/ and ~/.claude/tasks/).
- [SAFE]: No indicators of malicious patterns, such as prompt injection, hardcoded credentials, or unauthorized network operations, were detected. The skill includes best-practice advice, such as centralizing sensitive operations like 'git push' to a lead agent.
- [SAFE]: The skill defines a communication surface between agents via SendMessage. While this represents an architectural surface for data exchange, the instructions are structural and do not incorporate untrusted external data in an unsafe manner.
Audit Metadata