asana

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly instructs the agent to connect to Asana via Membrane and run actions such as "list-task-comments" and "list-tasks", which fetch user-generated comments and project data from an external Asana workspace that the agent will read and which could contain actionable instructions.