Skills
Navigate
skills/sonilo-ai/skills/video-to-sound/Gen Agent Trust Hub

video-to-sound

Pass

Audited by Gen Agent Trust Hub on Sep 29, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data that could contain malicious instructions.
  • Ingestion points: The skill accepts video_path, video_url, music_prompt, and sfx_prompt as inputs (SKILL.md).
  • Boundary markers: No specific delimiters or safety warnings for ignoring embedded instructions in the prompts or video metadata are defined.
  • Capability inventory: The skill uses Bash for command execution and mcp__sonilo__* tools for processing, alongside file system access for reading videos and writing output.
  • Sanitization: There is no evidence of sanitization or strict validation of the music_prompt or sfx_prompt inputs before they are passed to the backend service.
  • [COMMAND_EXECUTION]: The skill instructs the agent to use Bash to execute CLI commands such as sonilo video-to-sound and curl for API interactions.
  • [EXTERNAL_DOWNLOADS]: The skill is capable of fetching content from remote sources via the video_url parameter. It also references the installation of the sonilo and sonilo-cli packages from public registries (NPM and PyPI).
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 29, 2026, 11:04 AM