elevenlabs-sound-effects

Pass

Audited by Gen Agent Trust Hub on May 19, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns, prompt injections, or unauthorized data access were detected. The skill's behavior aligns with its stated purpose of sound effect generation.
  • [EXTERNAL_DOWNLOADS]: The skill references installation instructions and images hosted on the vendor's official GitHub repository and domain (inference.sh). These are trusted resources within the context of the skill's authorship.
  • [COMMAND_EXECUTION]: The skill uses the 'belt' CLI tool to execute sound generation tasks. This is restricted by the 'allowed-tools' configuration and represents the primary intended functionality of the skill.
  • [INDIRECT_PROMPT_INJECTION]: While the skill ingests user-provided text for sound generation, it lacks high-risk capabilities like sensitive file access or arbitrary network operations, which limits the potential impact of indirect prompt injection.
Audit Metadata
Risk Level
SAFE
Analyzed
May 19, 2026, 07:48 AM
Security Audit — agent-trust-hub — elevenlabs-sound-effects