video-edit

Pass

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: SAFEREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The installation instructions promote a one-liner command that downloads and executes a setup script directly from the author's GitHub repository: https://raw.githubusercontent.com/hoodini/ai-agents-skills/master/install.sh | bash.
  • [COMMAND_EXECUTION]: The skill automates video processing by orchestrating several CLI tools, including ffprobe for media analysis, ffmpeg for audio/video manipulation, and npx hyperframes for rendering the final output.
  • [EXTERNAL_DOWNLOADS]: The transcript editor webapp (index.html) dynamically loads the web-llm library from https://esm.run/@mlc-ai/web-llm to provide in-browser AI suggestions. This involves downloading machine learning models to the local browser cache via WebGPU.
  • [COMMAND_EXECUTION]: The script serve_review.py launches a local HTTP server on 127.0.0.1 to host the interactive review tool. This server is permitted to read project files and write the user's approved transcript revisions back to the filesystem.
  • [COMMAND_EXECUTION]: The skill maintains project metadata and usage history in the user's home directory at ~/.hyperframes-editor/projects.json to enable session persistence and the 'recent projects' feature in the web interface.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 17, 2026, 07:43 AM
Security Audit — agent-trust-hub — video-edit