azure-communication-callautomation-java

Pass

Audited by Gen Agent Trust Hub on Aug 9, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references the official azure-communication-callautomation Java SDK (version 1.6.0) from the com.azure organization, which is a trusted source.
  • [DATA_EXFILTRATION]: The skill facilitates network communication with Azure Communication Services endpoints (communication.azure.com) to manage call logic and media streaming, which is standard for cloud-based communication services.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources such as webhook callbacks, DTMF tones, and speech recognition results, creating a potential attack surface for indirect prompt injection.
  • Ingestion points: Inbound webhook request bodies processed by the handleCallback method and recognition results from DTMF or speech-to-text operations.
  • Boundary markers: No specific boundary markers or delimiters are demonstrated in the code snippets; the skill relies on the application logic to handle parsed events.
  • Capability inventory: The skill possesses capabilities to control calls (answer, transfer, hang up), play audio, and download recording files to the local filesystem using Paths.get("recording.mp4").
  • Sanitization: The skill utilizes the CallAutomationEventParser provided by the official Azure SDK to validate and parse incoming event data structures.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 9, 2026, 06:33 PM
Security Audit — agent-trust-hub — azure-communication-callautomation-java