elevenlabs
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied text for audio generation, which could be exploited to inject instructions or generate unintended content.\n
- Ingestion points: User-provided text strings for text-to-speech, sound effects, and music generation in
SKILL.md.\n - Boundary markers: No delimiters or protective instructions are used to separate user input from agent instructions.\n
- Capability inventory: Network access to ElevenLabs API and file system write operations for saving
.mp3files.\n - Sanitization: No input validation or sanitization is mentioned in the provided scripts or instructions.\n- [COMMAND_EXECUTION]: The skill guides the user to execute shell commands for media processing and workflow automation.\n
- Evidence: Usage of
ffmpegfor audio transcoding and execution of a local Python scripttools/voiceover.pyfor batch processing scenes.
Audit Metadata