openai-whisper
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill acts as a configuration and usage guide for the official OpenAI Whisper CLI tool, which is a reputable and widely-used software package.
- [COMMAND_EXECUTION]: The skill facilitates the execution of the whisper command for audio transcription. This local processing is the primary intended function and does not involve suspicious or hidden commands.
- [EXTERNAL_DOWNLOADS]: Installation is handled via official package managers (Homebrew) and models are retrieved from the developer's infrastructure. These references to official sources do not pose a security risk.
- [INDIRECT_PROMPT_INJECTION]: The skill transcribes external audio data into the agent's context. Ingestion points: audio files passed to the CLI (SKILL.md). Boundary markers: not specified in current instructions. Capability inventory: local command execution. Sanitization: not specified. This capability is expected for speech-to-text tools and relies on the underlying tool's integrity.
Audit Metadata