vendor-discovery-agent

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md explicitly instructs the agent to ingest and interpret open/public third‑party sources (e.g., Step 3 "Direct Research: Company websites, LinkedIn Sales Navigator, financial filings, Crunchbase, government registries like SAM.gov" and Step 6 "cross‑reference self‑reported data against third‑party sources") and to use those findings to score, shortlist, and drive next actions, which creates a clear pathway for indirect prompt injection from untrusted content.