elevenlabs-dialogue

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references installation instructions for the 'belt' CLI from the inference-sh GitHub repository and suggests adding the 'belt-sh/cli' package using npx. These resources are official components of the platform described in the skill.
  • [COMMAND_EXECUTION]: The skill documentation provides numerous examples of shell commands using the 'belt' CLI tool to run applications, log in, and list categories. These commands are restricted to the 'Bash(belt *)' tool scope as defined in the skill frontmatter.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied text to generate dialogue audio, creating a potential surface for indirect prompt injection.
  • Ingestion points: Text segments provided in the JSON input to the 'belt app run' command in SKILL.md.
  • Boundary markers: None present; the skill treats user-provided text as direct input for the audio generation process.
  • Capability inventory: Executes shell commands via the 'belt' CLI to interact with external APIs.
  • Sanitization: No explicit sanitization or filtering of the input text is described in the skill instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 05:53 PM
Security Audit — agent-trust-hub — elevenlabs-dialogue