elevenlabs
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external script files like VOICEOVER-SCRIPT.md and natural language prompts, creating a surface for indirect prompt injection where untrusted data could influence agent behavior.
- Ingestion points: Narration text is read from project files and audio prompts are provided by users.
- Boundary markers: No explicit delimiters are specified to isolate external content from instructions.
- Capability inventory: The skill performs network requests to ElevenLabs APIs, writes audio files to disk, and executes ffmpeg for processing.
- Sanitization: The instructions do not describe specific sanitization of input text before processing.
- [COMMAND_EXECUTION]: The skill utilizes ffmpeg for standard audio tasks such as format conversion and timing synchronization.
- [EXTERNAL_DOWNLOADS]: The skill relies on official and well-known libraries including the elevenlabs Python package and the remotion Node.js library for core functionality.
Audit Metadata