iblai-api-agent-session

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

• Third-party content exposure detected (high risk: 0.85). The required runtime workflow POSTs free-form metadata (arbitrary JSON key/values) from the caller into the chat turn, and the server-side runner appends it into the prompt for the LLM as a <CONTEXT METADATA>...</END CONTEXT METADATA> block (i.e., outsider-controlled request body content can become LLM context).