azure-speech-to-text
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill makes network requests to official Microsoft Azure AI Speech service endpoints to perform audio transcription. Microsoft is a well-known and trusted service provider.\n- [CREDENTIALS_UNSAFE]: The skill correctly identifies and uses environment variables (
AZURE_SPEECH_KEY,AZURE_SPEECH_REGION) for API authentication, avoiding the security risk of hardcoding sensitive secrets directly in the source code or documentation.\n- [INDIRECT_PROMPT_INJECTION]: \n - Ingestion points: The skill processes external audio and video files as input for the transcription engine via the
input_pathparameter.\n - Boundary markers: There are no explicit delimiters for the media content, which is standard for binary audio processing.\n
- Capability inventory: The tool performs local file read operations to access audio assets and network POST operations to the Azure REST API.\n
- Sanitization: The skill transcribes the content as-is; security depends on the downstream application handling the resulting text transcript, where standard AI guardrails are typically applied.
Audit Metadata