transcribe
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides a standard interface for audio transcription. It uses the official
openaiPython library and communicates exclusively with OpenAI's infrastructure, which is a recognized trusted service. The bundled scripttranscribe_diarize.pyuses standard libraries for file and argument handling without any malicious execution patterns. - [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes untrusted audio data that could contain spoken instructions designed to manipulate the agent's behavior. \n
- Ingestion points: Audio files are read by the
scripts/transcribe_diarize.pyscript from paths provided via command-line arguments. \n - Boundary markers: The transcription results are output as raw text or JSON; the skill does not wrap this output in protective delimiters or provide instructions to the agent to treat the content as untrusted. \n
- Capability inventory: The skill includes file system read access and network access to the OpenAI API. \n
- Sanitization: No content filtering or sanitization is performed on the audio input or the transcribed output.
Audit Metadata