senior-data-engineer

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

• Third-party content exposure detected (medium risk: 0.30). In the required “real-time streaming” workflow, the pipeline reads JSON event records from an outsider-controllable Kafka topic (“user-events”) via spark.readStream.format("kafka")...load() and then parses that free-text value field with from_json(...), so injected content in the topic becomes LLM-visible input at runtime.