AgentSkillsCN

gem

使用 Google Gemini 进行多模态 AI 处理与图像生成。适用于分析 PDF、图片、视频、YouTube 链接及其他大型文档。还可生成 Nano Banana Pro 图像。非常适合需要从需要视觉或多模态理解的文件中提取信息，或根据文本提示生成图像时使用。

SKILL.md
--- frontmatter
name: gem
description: Multimodal AI processing and image generation using Google Gemini. Use for analyzing PDFs, images, videos, YouTube links, and other large documents. Also generates images with Nano Banana Pro. Ideal when you need to extract information from files that require vision or multimodal understanding, or generate images from text prompts.

Gemini Multimodal Tool

Use the ai-gem CLI tool for multimodal AI processing and image generation via Google's Gemini API.

Usage

bash
# Text queries
ai-gem "Write a haiku about Python programming"

# Analyze documents
ai-gem "Summarize this document" document.pdf

# Analyze images
ai-gem "What's in this image?" photo.jpg

# Process YouTube videos
ai-gem "Create a 5-point summary" "https://youtu.be/VIDEO_ID"

# Compare multiple files
ai-gem "Compare these files" file1.pdf file2.png

# Web search
ai-gem "Current AI news" --search

# Generate images (uses Nano Banana Pro by default)
ai-gem --image "A cute robot reading a book in a cozy library"
ai-gem --image "A landscape at sunset" --aspect-ratio 16:9
ai-gem --image "A cat wearing a hat" -o cat.png
ai-gem --image "Edit this to add sunglasses" reference.jpg

# Use alternative image model
ai-gem --image "A blue triangle" -m gemini-2.5-flash-image

Image Generation Options

  • •--image / -i: Generate an image instead of text
  • •--output / -o: Output file path (auto-generated if omitted)
  • •--aspect-ratio / -a: Aspect ratio (1:1, 9:16, 16:9, etc.)
  • •--model / -m: Override model (default: nano-banana-pro-preview)
  • •Attachments serve as reference images for editing

Requirements

  • •GEMINI_API_KEY environment variable must be set
  • •The hamel package must be installed: pip install hamel

Supported Input Types

  • •PDFs
  • •Images (PNG, JPEG, GIF, WebP)
  • •Videos (MP4, etc.)
  • •YouTube URLs
  • •Plain text files
  • •Multiple files for comparison