audio-gen
Pass
Audited by Gen Agent Trust Hub on Jun 22, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: No hardcoded credentials or unauthorized data exfiltration patterns were detected. The skill correctly instructs the user to manage the 'ELEVENLABS_API_KEY' using environment variables, which is a standard security practice.
- [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection (Category 8). It ingests untrusted data from users, including text for TTS and audio/video files for cloning and dubbing. The instructions lack explicit boundary markers or sanitization requirements to prevent the agent from obeying instructions that might be embedded within the processed content.
- Ingestion points: User-provided text prompts, referenced audio (WAV/MP3), and video (MP4) files.
- Boundary markers: None identified in the skill documentation.
- Capability inventory: The skill utilizes the 'elevenlabs-mcp' server which performs network operations to the ElevenLabs API and handles local file read/write operations for media processing.
- Sanitization: No specific sanitization or validation steps for external content are defined in the instructions.
Audit Metadata