plagiarism-checker

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

• Third-party content exposure detected (high risk: 1.00). The workflow explicitly instructs the agent to perform WebSearch on extracted phrases and to WebFetch lyrics pages from public websites (see Step 3 "Web Search" and Step 4 "WebFetch the lyrics page"), requiring the agent to read untrusted third‑party content and use it to drive plagiarism risk decisions—creating a clear vector for indirect prompt injection.