text-to-speech

Pass

Audited by Gen Agent Trust Hub on Sep 21, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references installation scripts and documentation from the inference.sh official GitHub repository and provides links to well-known AI services including Inworld AI and ElevenLabs.
  • [COMMAND_EXECUTION]: The skill utilizes the 'belt' CLI tool to execute text-to-speech tasks. All tool usage is restricted to the 'belt' command scope via the allowed-tools configuration in the YAML frontmatter.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided text for conversion into speech, which represents a potential injection surface.
  • Ingestion points: Text inputs passed via the --input flag to the 'belt app run' command in SKILL.md.
  • Boundary markers: Data is passed using structured JSON formats, helping to distinguish between input text and execution parameters.
  • Capability inventory: The skill is scoped specifically to speech synthesis operations using the belt CLI; it does not possess file-write or generalized network capabilities.
  • Sanitization: Inputs are handled by the underlying speech synthesis APIs which typically implement their own safety and content filtering layers.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 21, 2026, 08:14 PM
Security Audit — agent-trust-hub — text-to-speech