multimodal-whisper
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a documentation resource for OpenAI's Whisper model.
- [EXTERNAL_DOWNLOADS]: Includes instructions to install official packages like
openai-whisperandfaster-whisperfrom standard registries. It also references well-known research and source code repositories. - [COMMAND_EXECUTION]: Provides standard command-line examples for transcribing audio and installing system dependencies (e.g.,
ffmpegviaapt,brew, orchoco). These are typical for developer tooling and occur in the context of user-initiated setup. - [DATA_EXPOSURE]: No patterns of unauthorized sensitive data access, hardcoded credentials, or exfiltration were detected.
Audit Metadata