byted-mediakit-audio

Pass

Audited by Gen Agent Trust Hub on Nov 5, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill facilitates audio media processing (probing, transcoding, and voice activity detection) using a dedicated command-line interface, mediakit-cli, as defined in its metadata.
  • [COMMAND_EXECUTION]: Shell access is requested to execute mediakit-cli commands. The instructions provide explicit guidance on safely formatting arguments, particularly for JSON payloads, to ensure correct shell execution without user input interpolation risks.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted media URLs (audio_url, video_url). The instructions include strict boundary rules: the agent must only pass user-provided values, must not fabricate parameters, and must clarify with the user if mandatory information is missing. This significantly reduces the surface for injection via parameter manipulation.
  • [SAFE]: The skill references vendor-specific protocols (vod://, tos://) for Volcengine's media and storage services, which is consistent with the skill's stated purpose and author context.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 5, 2026, 09:02 AM