azure-ai-agents-persistent-java
Pass
Audited by Gen Agent Trust Hub on Feb 14, 2026
Risk Level: LOWEXTERNAL_DOWNLOADSPROMPT_INJECTIONSAFE
Full Analysis
- [EXTERNAL_DOWNLOADS] (INFO): References the 'com.azure:azure-ai-agents-persistent' Maven dependency. This is an official library from a trusted source (Microsoft/Azure).
- [CREDENTIALS_UNSAFE] (SAFE): Properly implements DefaultAzureCredentialBuilder, which avoids sensitive credential exposure.
- [PROMPT_INJECTION] (LOW): Detected an indirect prompt injection surface. (1) Ingestion points: client.createMessage in SKILL.md. (2) Boundary markers: Absent. (3) Capability inventory: Interactions with Azure AI service for agent runs and message management. (4) Sanitization: Absent. Risk is restricted to influence over the agent reasoning and state (LOW tier).
Audit Metadata