dispatching-agent-teams

Pass

Audited by Gen Agent Trust Hub on Apr 25, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns detected. The skill acts as an orchestrator for existing agent and skill definitions.
  • [SAFE]: The skill implements verification steps to ensure only officially installed agents and skills are resolved, mitigating the risk of unauthorized component execution (SKILL.md, Phase 1).
  • [SAFE]: In Sequential Mode, the protocol recommends using worktree isolation when agents need to write files, which is a security best practice for multi-agent environments (references/spawning-protocol.md).
  • [SAFE]: The 'War Room' protocol explicitly restricts sub-agents from using tools, reading files, or executing commands, which effectively mitigates the risk of automated actions or exfiltration during debate phases (references/spawning-protocol.md).
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 25, 2026, 11:10 AM