Watch a video (URL or local path). Downloads with yt-dlp, extracts auto-scaled frames with ffmpeg, pulls the transcript from captions (or local WhisperX / cloud Whisper fallback), and hands the result to the agent so it can answer questions about what's in the video. With a Gemini API key, Google's agentic video model watches the full video instead.
Install
npx skills add bradautomates/claude-video --skill watchnpx skills add bradautomates/claude-video --skill watch -a claude-code -gnpx skills add bradautomates/claude-video --skill watch -a codex -ggemini skills install https://github.com/bradautomates/claude-video.git --path skills/watchgh skill install bradautomates/claude-video watchgit clone --depth 1 https://github.com/bradautomates/claude-video /tmp/bradautomates__claude-video
mkdir -p ~/.claude/skills ~/.agents/skills
cp -R /tmp/bradautomates__claude-video/skills/watch ~/.claude/skills/watch # Claude Code
cp -R /tmp/bradautomates__claude-video/skills/watch ~/.agents/skills/watch # Codex, Cursor, Gemini CLI, Copilot, OpenCode