azure-eventhub-dotnet
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSMETADATA_POISONING
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides code to ingest data from Azure Event Hubs via
EventProcessorClient.ProcessEventAsync(SKILL.md, line 146). This creates an attack surface for indirect prompt injection where malicious instructions embedded in event bodies could influence agent behavior. 1. Ingestion point:args.Data.EventBody. 2. Boundary markers: Absent. 3. Capability inventory: IncludesEventHubProducerClientfor sending messages andBlobContainerClientfor storage operations. 4. Sanitization: None provided in snippets. - [EXTERNAL_DOWNLOADS]: The skill specifies the installation of several official Microsoft NuGet packages (Azure.Messaging.EventHubs, Azure.Identity, etc.) from a well-known service (SKILL.md, lines 14-25).
- [METADATA_POISONING]: The YAML frontmatter contains a
risk: criticalfield (SKILL.md, line 4) which is misleading as the skill primarily contains standard SDK documentation and best practices.
Audit Metadata