Skills
skills/inference-sh/skills/elevenlabs-stt/Gen Agent Trust Hub

elevenlabs-stt

Pass

Audited by Gen Agent Trust Hub on Apr 23, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references configuration and installation documentation for the infsh CLI from the author's repository on GitHub.
  • [COMMAND_EXECUTION]: The skill utilizes the infsh command-line tool to execute audio processing tasks. The execution scope is restricted to the infsh tool via the allowed-tools configuration in the skill metadata.
  • [PROMPT_INJECTION]: The skill ingests untrusted data in the form of audio URLs and text strings for transcription and alignment tasks, which constitutes a surface for indirect prompt injection.
  • Ingestion points: Untrusted audio URLs and text payloads are passed to the infsh command via JSON input in SKILL.md examples.
  • Boundary markers: Data is encapsulated within structured JSON objects.
  • Capability inventory: The skill is capable of executing shell commands via the infsh CLI tool.
  • Sanitization: No specific sanitization logic is visible in the markdown instructions, relying on the underlying CLI and agent implementation for input handling.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 23, 2026, 07:52 AM