tavus-cvi-knowledge
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides legitimate documentation for the Tavus API. All network operations target the vendor's domain (tavusapi.com). API keys are correctly managed using safe placeholders ("YOUR_API_KEY").
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external data (PDFs, websites) via the
document_urlparameter. This introduces an inherent surface for indirect prompt injection common to Retrieval-Augmented Generation (RAG) systems. - Ingestion points:
document_urlfield inSKILL.md. - Boundary markers: None defined in the skill documentation.
- Capability inventory: The skill interacts with the Tavus API to provide context to personas but does not execute local commands.
- Sanitization: Not specified within the skill instructions.
Audit Metadata