plg-ai-funnel

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 in "Stage 2: Documentation Scan" and the Quick Audit Workflow that the agent should "scan your docs, help center, marketing pages" (public web content) and then use those extracts to drive feature matching and recommendations, meaning untrusted third‑party pages could materially influence agent actions.