dust-call-agent

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md explicitly says the agent can "read a slack thread, a notion URL, a google drive document..." and use that content to drive actions (e.g., create issues, post messages), so untrusted/user-generated third-party content can be ingested and materially influence agent behavior.