ai-rag-patterns

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

• Third-party content exposure detected (medium risk: 0.65). Outsider-authored free text can enter the LLM context via the RAG retrieval path: retrieved chunk c['text'] (sourced from documents loaded from external inputs like web pages/URLs or other users’ content) is concatenated into the prompt in _build_prompt() / _format_context() and thus fed to the agent LLM.