dispatching-parallel-agents
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is a documentation-based guide that outlines best practices for agent orchestration and parallel task management.- [INDIRECT_PROMPT_INJECTION]: The skill describes a workflow where an agent processes external data (such as test failure reports) to delegate tasks to sub-agents. While this creates a potential surface for indirect prompt injection if sub-agent inputs are derived from untrusted sources, the skill explicitly recommends mitigation strategies such as providing isolated contexts, defining clear goals, and setting behavioral constraints to limit the scope of sub-agents.
Audit Metadata