skills/agisota/openmontage/doubao-tts/Gen Agent Trust Hub

doubao-tts

Pass

Audited by Gen Agent Trust Hub on Aug 26, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill interacts with official Bytedance Volcengine endpoints at openspeech.bytedance.com for submitting and querying text-to-speech jobs.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: Configuration instructions specify using the DOUBAO_SPEECH_API_KEY environment variable. The skill includes explicit safety instructions for the agent to never log, print, or store this key in project artifacts.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data via the text parameter for narration.
  • Ingestion points: User-provided text parameter used in TTSSelector and DoubaoTTS tools within SKILL.md.
  • Boundary markers: None explicitly defined in the prompt interpolation, but the skill mandates a human-in-the-loop approval workflow for audio samples.
  • Capability inventory: The skill performs file writes to output_path and network requests to openspeech.bytedance.com.
  • Sanitization: No specific sanitization or filtering of the input text is described in the instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 26, 2026, 06:33 PM