diarize-api
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEMETADATA_POISONINGEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [METADATA_POISONING]: The skill documentation exhibits significant internal contradictions. The name and API endpoints refer to 'Speaker Diarization' (audio processing), but the description and detailed instructions refer to 'image generation/editing', 'image size' enums, and variations. This discrepancy appears to be a non-malicious documentation error.
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to fetch reference documentation from the vendor's official domain at 'https://www.pixazo.ai/models/diarize-api.md' and mentions a command to install additional skills from the vendor's GitHub organization.
- [COMMAND_EXECUTION]: The instructions include standard shell commands (curl) and file system operations (chmod 600) for managing the user's API key and interacting with the service provider's API gateway.
Audit Metadata