tahr-test-ai-agents

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

• Third-party content exposure detected (high risk: 0.75). The required workflow is a security review that traces and probes AI/RAG surfaces, which implies the agent will ingest attacker-controlled free-form content (e.g., uploaded documents, fetched URLs, chat messages) into the LLM context for indirect prompt injection testing.