Gen Agent Trust Hub

The skill is designed for product management tasks within a GitHub repository but contains instructions that could lead to unauthorized privilege escalation. Specifically, it encourages the agent to execute administrative commands (sudo) during the onboarding process. Additionally, the skill's reliance on reading external data like pull request comments and issue descriptions creates a surface for indirect prompt injection, as the agent has the capability to approve pull requests and modify repository content based on that external input.

The provider classified the analyzed snapshot as HIGH risk. This result applies to the content available at audit time.