elevenlabs-narration
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOWPROMPT_INJECTION
Full Analysis
- [Indirect Prompt Injection] (LOW): The skill is designed to process external text for audio synthesis, which creates a standard surface for indirect prompt injection. * Ingestion points: The 'text' parameter in the 'generateNarration' function within SKILL.md. * Boundary markers: No delimiters or 'ignore' instructions are provided to separate the input content from instructions. * Capability inventory: The skill utilizes the 'elevenlabs' package to make network requests to the ElevenLabs API (api.elevenlabs.io). * Sanitization: No text sanitization or input validation is implemented in the code samples.
Audit Metadata