azure-eventhub-py

Pass

Audited by Gen Agent Trust Hub on Feb 14, 2026

Risk Level: LOWPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION] (LOW): Indirect Prompt Injection Surface detected. The skill is designed to ingest and process external data from Azure Event Hubs, which can contain untrusted instructions. * Ingestion points: The on_event callback in SKILL.md and references/acceptance-criteria.md extracts external data via event.body_as_str(). * Boundary markers: None are present to delimit untrusted data from agent instructions. * Capability inventory: The provided code demonstrates low-risk capabilities including printing to console (print), state management (update_checkpoint), and sending events (send_batch). No arbitrary command execution or file system writes are present. * Sanitization: No sanitization or validation of the event body is performed before display.
  • [EXTERNAL_DOWNLOADS] (LOW): The skill requires installation of azure-eventhub, azure-identity, and azure-eventhub-checkpointstoreblob-aio. These are official packages from Microsoft/Azure. Per [TRUST-SCOPE-RULE], these findings are downgraded to LOW/INFO as they originate from a trusted organization.
  • [SAFE] (SAFE): No evidence of hardcoded credentials, malicious obfuscation, or unauthorized data exfiltration was found. The use of DefaultAzureCredential follows cloud security best practices.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 14, 2026, 04:59 PM