langfuse-data-handling

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

• Third-party content exposure detected (high risk: 0.70). The required workflow calls the Langfuse API at runtime and ingests trace/observation content returned from that external service (which is outsider-authored user/trace data), placing it into the program’s in-memory/LLM-relevant context when subsequently logged/serialized as free text.