discovery-interview-prep

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly asks the user to paste or provide third-party materials in "Step 0: Gather Context" (e.g., "copy competitor materials, customer reviews (G2, Capterra), or community discussions (Reddit, forums)" and "You can paste this content directly"), so the agent will ingest untrusted, user-generated public web content and use it in its workflow.