elevenlabs-dialogue
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references the
beltCLI tool and provides installation links to the inference-sh GitHub repository and the inference.sh domain. These resources are standard for the advertised functionality of the service. - [COMMAND_EXECUTION]: The skill uses the
beltcommand-line utility within a Bash environment to execute remote AI applications. The execution scope is limited to thebelttool as defined in the frontmatter. - [INDIRECT_PROMPT_INJECTION]: The skill provides an interface for processing user-supplied text through an external API.
- Ingestion points: User-provided JSON payloads containing dialogue text (SKILL.md).
- Boundary markers: Instructions use structured JSON fields (
"text","voice") to separate instructions from data. - Capability inventory: Uses
Bash(belt *)to send data to the remote inference service. - Sanitization: The skill assumes the underlying CLI tool and API handle sanitization of the generated audio content.
Audit Metadata