kyc-doc-parse

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly ingests untrusted onboarding documents supplied by the applicant (see SKILL.md: "Input is untrusted. Onboarding documents are supplied by the applicant"), and the agent is required to read and extract that user-provided content to produce structured KYC outputs that feed downstream decisions, so third‑party content can materially influence behavior.