azure-eventhub-java
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides patterns for consuming data from external event streams, creating a surface where malicious instructions could be embedded in the data being processed.
- Ingestion points: Data enters the agent's context via
event.getBodyAsString()and properties in theEventProcessorClientandreceiveFromPartitionsections ofSKILL.md. - Boundary markers: The implementation examples do not use explicit delimiters or "ignore instructions" markers to wrap incoming event data.
- Capability inventory: The skill includes capabilities to log data to the console (
System.out.println) and send data back to the Event Hub (producer.send), creating a potential feedback loop. - Sanitization: The code examples lack validation or sanitization of the event bodies before they are processed by the application logic.
- [EXTERNAL_DOWNLOADS]: The skill requires official Java libraries for Azure Event Hubs and Azure Storage.
- Evidence: Maven dependencies for
com.azure:azure-messaging-eventhubsandcom.azure:azure-messaging-eventhubs-checkpointstore-blobare specified inSKILL.md. - Note: These are official SDKs from a well-known service provider and are used strictly for the skill's primary purpose of event processing.
Audit Metadata