Gen Agent Trust Hub

The skill coordinates parallel workers using subagents and aggregates their results into a report. It includes features for managing workload isolation and dynamically configuring agent types based on reasoning levels. While it follows best practices by instructing the agent to summarize rather than dump raw data, the aggregation of untrusted worker output represents an indirect prompt injection surface.

The provider classified the analyzed snapshot as NONE risk. This result applies to the content available at audit time.