Gemini System — Multimodal File Processing

Gemini CLI is specialized for multimodal file processing (PDF, video, audio, image).

Detailed rules: .claude/rules/gemini-delegation.md Research/codebase analysis: Use general-purpose subagent (Opus) instead — Opus now supports 1M context.

Multimodal File Processing

Extract content from PDF, video, audio, and image files.

# PDF
gemini -p "Extract: {what to extract}" < /path/to/file.pdf 2>/dev/null

# Video
gemini -p "Summarize: key concepts, timestamps" < /path/to/video.mp4 2>/dev/null