azure-ai-transcription-py

Warn

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill recommends installing a package named azure-ai-transcription. This is not the official Microsoft Azure library for speech services (which is typically azure-cognitiveservices-speech), representing an unverifiable dependency from an unknown source.
  • [REMOTE_CODE_EXECUTION]: The provided code snippets import and execute methods from the azure-ai-transcription library. Running code from an unverifiable package can lead to the execution of malicious scripts.
  • [PROMPT_INJECTION]: The skill contains an attack surface for indirect prompt injection.
  • Ingestion points: The skill ingests untrusted audio data from external URLs provided in the content_urls parameter or local audio files.
  • Boundary markers: There are no instructions or delimiters implemented to prevent the agent from following commands that might be embedded in the transcribed text.
  • Capability inventory: The skill outputs transcribed text directly into the agent's context, which could influence downstream tasks.
  • Sanitization: There is no evidence of sanitization, filtering, or validation of the transcription output before it is processed.
  • [SAFE]: The skill references the official Azure SDK GitHub repository for testing criteria and uses standard environment variable patterns for credential management.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 1, 2026, 12:34 AM