AI Commands

The floimg ai command group provides AI-powered text generation, image analysis, and image editing using Gemini.

Prerequisites

Install the Google AI plugin:

npm install -g @teamflojo/floimg-google

Set your API key:

export GOOGLE_AI_API_KEY=your-api-key

Get your API key from Google AI Studio.

---

Text Generation

Generate text using Gemini AI.

floimg ai text --prompt "Write 3 creative prompts for a fantasy landscape"

Options

Option	Required	Description
--prompt <text>	Yes	Text prompt for generation
--api-key <key>	No	API key (or use GOOGLE_AI_API_KEY env)
--model <name>	No	Model name (default: gemini-2.0-flash)
--output-format <fmt>	No	text or json (default: text)
--context <text>	No	Additional context
--out <path>	No	Save output to file

Examples

# Simple text generation
floimg ai text --prompt "Describe a cozy coffee shop in one paragraph"

# Generate structured JSON
floimg ai text \
  --prompt "Generate 3 image prompts for product photos" \
  --output-format json

# Save to file
floimg ai text --prompt "Write alt text for a product image" --out alt.txt

# Use specific model
floimg ai text --prompt "Complex analysis task" --model gemini-2.0-flash

---

Vision Analysis

Analyze images using Gemini Vision.

floimg ai vision --image photo.jpg --prompt "What objects are in this image?"

Options

Option	Required	Description
--image <path>	Yes	Input image path
--prompt <text>	Yes	Analysis prompt
--api-key <key>	No	API key (or use GOOGLE_AI_API_KEY env)
--model <name>	No	Model name (default: gemini-2.0-flash)
--output-format <fmt>	No	text or json (default: text)
--out <path>	No	Save output to file

Examples

# Describe an image
floimg ai vision --image landscape.jpg --prompt "Describe this scene"

# Extract structured data
floimg ai vision \
  --image product.png \
  --prompt "List all products visible with prices" \
  --output-format json

# Generate alt text
floimg ai vision \
  --image hero.jpg \
  --prompt "Write accessible alt text for this image" \
  --out alt.txt

# Content analysis
floimg ai vision \
  --image photo.jpg \
  --prompt "Is this image appropriate for a family website?"

---

Image Editing

Edit images using Gemini’s native image generation (Nano Banana).

floimg ai edit --image input.jpg --prompt "Make the sky more vibrant" --out edited.jpg

Options

Option	Required	Description
--image <path>	Yes	Input image path
--prompt <text>	Yes	Edit instructions
--out <path>	Yes	Output image path
--api-key <key>	No	API key (or use GOOGLE_AI_API_KEY env)
--model <name>	No	Model (default: gemini-2.5-flash-image)
--pre-prompt <text>	No	Instructions prepended to prompt
--aspect-ratio <ratio>	No	Output aspect ratio (default: 1:1)
--image-size <size>	No	Output resolution: 1K, 2K, or 4K (default: 1K)
--grounding	No	Enable Google Search grounding

Aspect Ratios

Supported: 1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9

Examples

# Simple edit
floimg ai edit \
  --image photo.jpg \
  --prompt "Remove the background and replace with white" \
  --out clean.jpg

# Style transfer with 4K output
floimg ai edit \
  --image portrait.jpg \
  --prompt "Convert to watercolor painting style" \
  --image-size 4K \
  --out watercolor.jpg

# Widescreen output
floimg ai edit \
  --image room.jpg \
  --prompt "Add a potted plant in the corner" \
  --aspect-ratio 16:9 \
  --out decorated.jpg

# Use grounding for real-time data
floimg ai edit \
  --image infographic.jpg \
  --prompt "Update the chart with current stock prices for AAPL" \
  --grounding \
  --out updated.jpg

Models

Model	Description
gemini-2.5-flash-image	Fast, high-volume (default)
gemini-3-pro-image-preview	Higher quality, better text rendering

---

Image Generation

Generate images from text prompts using Gemini.

floimg ai generate --prompt "A cozy coffee shop interior" --out coffee-shop.png

Options

Option	Required	Description
--prompt <text>	Yes	Image description
--out <path>	Yes	Output image path
--api-key <key>	No	API key (or use GOOGLE_AI_API_KEY env)
--model <name>	No	Model (default: gemini-2.5-flash-image)
--aspect-ratio <ratio>	No	Output aspect ratio (default: 1:1)
--image-size <size>	No	Output resolution: 1K, 2K, or 4K (default: 1K)
--grounding	No	Enable Google Search grounding

Examples

# Generate an image
floimg ai generate \
  --prompt "A minimalist workspace with natural lighting" \
  --out workspace.png

# Widescreen cinematic image
floimg ai generate \
  --prompt "A photorealistic mountain lake at sunrise, mist rising from the water" \
  --aspect-ratio 21:9 \
  --image-size 4K \
  --out landscape.png

# Specific style
floimg ai generate \
  --prompt "Isometric 3D illustration of a tiny house, pastel colors, soft shadows" \
  --out house.png

---

Pipeline Examples

Combine AI commands with other floimg operations:

Analyze and Enhance

# 1. Analyze the image
floimg ai vision \
  --image photo.jpg \
  --prompt "Suggest one improvement for this photo" \
  --out suggestion.txt

# 2. Apply the suggestion
floimg ai edit \
  --image photo.jpg \
  --prompt "$(cat suggestion.txt)" \
  --out enhanced.jpg

Generate and Transform

# 1. Generate an image
floimg ai generate \
  --prompt "Abstract geometric pattern" \
  --out pattern.png

# 2. Resize for different uses
floimg resize pattern.png 1920x1080 --out pattern-desktop.png
floimg resize pattern.png 1080x1920 --out pattern-mobile.png

Vision to Edit Pipeline

# 1. Analyze the current state
floimg ai vision \
  --image room.jpg \
  --prompt "What furniture is missing to make this room complete?" \
  --out analysis.txt

# 2. Use vision output to guide edit
floimg ai edit \
  --image room.jpg \
  --prompt "Add the following: $(cat analysis.txt)" \
  --out furnished.jpg

---

Environment Variables

Variable	Description
GOOGLE_AI_API_KEY	Google AI API key for Gemini

See Also

• Google AI Plugin - Gemini SDK documentation
• xAI Plugin - Grok text and vision
• Transform Command - Non-AI image transforms
• MCP Tools - AI tools for MCP agents