elevenlabs-tts
Pass
Audited by Gen Agent Trust Hub on May 13, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides links to installation instructions for the
beltCLI tool from the developer's public GitHub repository. - [COMMAND_EXECUTION]: The skill uses the
beltCLI to interact with the ElevenLabs synthesis engine via theinference.shplatform. - [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it processes user-provided text for audio generation.
- Ingestion points: User-supplied text strings passed to the
textparameter in thebelt app runcommand inSKILL.md. - Boundary markers: External data is contained within a structured JSON object passed as a command argument.
- Capability inventory: Execution of subprocess commands via the
beltCLI tool as described inSKILL.md. - Sanitization: No explicit sanitization or filtering of user text is documented within the skill instructions.
Audit Metadata