evidence-ledger-research

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

• Third-party content exposure detected (medium risk: 0.65). The required workflow is described as “web research, document QA, PDF/table extraction” and “inputs: user questions, documents, URLs, data tables, extracted facts,” which implies runtime ingestion of outsider-authored free text from fetched public web pages or user-provided documents/URLs into the LLM context.