clay-reference-architecture

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

• Third-party content exposure detected (high risk: 0.80). This skill explicitly ingests external webhook payloads (see SKILL.md Step 3 and the webhook handler in references/implementation-guide.md) and directly uses fields like data.icp_score and data.personalized_intro to trigger actions (pushToCRM, addToInstantly), so untrusted third-party content can materially influence tool use and decisions.