메뉴
Google Gemini 2.5 Flash Image (Nano Banana) model for generating, editing, and creating AI-powered visual content including infographics, product photos, illustrations, and more. Use when the user asks about nanobana image generator.
기본적으로 소스를 먼저 확인하는 Prompt가 선택됩니다. 직접 명령으로 전환하거나 로컬 사본을 다운로드할 수도 있습니다.
설치 여부를 결정하기 전에 SKILL.md와 SkillsMP에 표시된 보조 파일을 읽어 보세요.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
직접 명령은 검토 Prompt를 거치지 않습니다. 실행하기 전에 소스를 확인하세요.
npx skills add https://github.com/agrivisionai/claude-skills --skill nanobana-image-generator명령은 한 줄로 유지됩니다. 복사하기 전에 가로로 스크롤해 전체 내용을 확인하세요.
로컬 사본을 원하시나요? SkillsMP에서 현재 제공할 수 있는 파일을 다운로드하세요.
Weather alerts with specific, actionable recommendations for farm operations. Use when the user asks about actionable weather alerts.
Validate equipment performance against marketing claims and specifications. Use when the user asks about actual vs promised validator.
Optimize critical operations timing for weather conditions and operational efficiency. Use when the user asks about critical timing optimizer.
SKILL.md 표시 중
SOC 직업 분류 기준
| name | nanobana-image-generator |
| description | Google Gemini 2.5 Flash Image (Nano Banana) model for generating, editing, and creating AI-powered visual content including infographics, product photos, illustrations, and more. Use when the user asks about nanobana image generator. |
Google Gemini 2.5 Flash Image (Nano Banana) model for generating, editing, and creating AI-powered visual content including infographics, product photos, illustrations, and more.
Provide farmers and agricultural content creators with AI-powered image generation capabilities using Google's Nano Banana model. This skill enables creation of professional infographics, product mockups, educational visuals, and marketing assets for agricultural documentation, presentations, and promotional materials.
Creating high-quality agricultural visual content is expensive and time-consuming. Hiring designers for infographics, product photos, and illustrations costs thousands annually. Farmers need professional visuals for grant applications, market materials, educational content, and social media, but lack design skills and budgets. This skill provides free AI-powered image generation for agricultural contexts.
Setup API Connection
Generate Simple Images
Create Complex Infographics
Edit Existing Images
Ground with Real-Time Data
Generate Infographics
Create Product Mockups
Edit Marketing Materials
Generate Educational Content
Description: Generate image from text description using Nano Banana model.
Parameters:
prompt (string, required): Natural language description of desired imageaspect_ratio (string, optional): Aspect ratio ("1:1", "16:9", "9:16", "21:9", etc.) Default: "1:1"resolution (string, optional): Image resolution ("1K", "2K", "4K") Default: "1K"model (string, optional): Model to use ("gemini-2.5-flash-image" or "gemini-3-pro-image-preview") Default: "gemini-2.5-flash-image"use_search (boolean, optional): Enable Google Search grounding Default: falseReturns: Generated image data with file path and metadata
Description: Edit existing image using text prompts.
Parameters:
image_path (string, required): Path to reference image to editprompt (string, required): Edit instructions (add, remove, modify elements)preserve_style (boolean, optional): Maintain original style, lighting, composition Default: trueReturns: Edited image with file path
Description: Create multi-turn conversation for iterative image refinement.
Parameters:
initial_prompt (string, required): Starting promptiterations (integer, optional): Number of refinement iterations Default: 3response_modalities (array, optional): Response types ("text", "image") Default: ["text", "image"]Returns: Conversation history with all generated images
Description: Generate multiple images in batch.
Parameters:
prompts (array, required): List of image generation promptscommon_config (object, optional): Shared configuration (aspect ratio, resolution) for all imagesReturns: Array of generated images with metadata
# Google AI API
GOOGLE_API_KEY=your_google_api_key_here
# Model Configuration
DEFAULT_MODEL=gemini-2.5-flash-image
DEFAULT_ASPECT_RATIO=1:1
DEFAULT_RESOLUTION=2K
# Search Configuration
ENABLE_GOOGLE_SEARCH=false
SEARCH_REGION=US
# Output Configuration
OUTPUT_DIR=outputs/images
IMAGE_FORMAT=png
SAVE_METADATA=true
# Image Generation
MAX_REFERENCE_IMAGES=14
ENABLE_THINKING_MODE=true
# File Handling
AUTO_OPEN_IMAGES=true
LOG_GENERATED_IMAGES=true
from nanobana_generator import NanoBananaGenerator
generator = NanoBananaGenerator()
prompt = """
Create a 3-panel infographic explaining crop rotation benefits for farmers.
Panel 1 (left): Show three-year crop rotation cycle with diagrams of corn, beans, wheat in sequence. Include crop rotation benefits: pest break, nutrient balance, soil health improvement.
Panel 2 (center): Show soil nitrogen levels before and after rotation with bar charts. Include legume nitrogen fixation visualization showing root nodules adding nitrogen.
Panel 3 (right): Show yield improvement data with line graph comparing rotated vs non-rotated fields over 5 years. Include economic benefit calculation.
Style: Professional infographic style with data visualizations, charts, and diagrams. Use agricultural color palette (green #2D7D2F, brown #8B4513, sky blue #87CEEB, yellow #FFD700). Include title "Crop Rotation Benefits" at top, labeled sections, and legend.
"""
image = generator.generate_image(
prompt=prompt,
aspect_ratio="16:9",
resolution="2K"
)
print(f"Generated infographic: {image['file_path']}")
from nanobana_generator import NanoBananaGenerator
generator = NanoBananaGenerator()
prompt = """
Professional product photograph of organic heirloom tomato seed packets on rustic wooden crate surface.
Composition:
- Center frame: 3 seed packets arranged in triangular composition
- Packets: Front-facing with clear variety names (Cherokee Purple, Brandywine, Mortgage Lifter)
- Background: Weathered wood texture with natural lighting
- Details: Sharp focus on seed packet text, visible seed shapes inside
Lighting: Studio three-point softbox setup, warm color temperature (5500K), subtle highlights on packet edges.
Camera: 50mm lens, f/2.8 aperture for shallow depth of field, product-only focus.
Style: Ultra-realistic, e-commerce quality, sharp focus, natural shadows.
"""
image = generator.generate_image(
prompt=prompt,
aspect_ratio="4:3",
resolution="2K"
)
print(f"Generated product photo: {image['file_path']}")
from nanobana_generator import NanoBananaGenerator
generator = NanoBananaGenerator()
# Load existing product photo
reference_image = "existing_product_photo.png"
prompt = """
Update the background of this product photo to match our farm brand identity:
1. Change background from plain white to gradient using brand colors: dark green (#1B4D3E) to light green (#5D8A4C)
2. Add subtle farm scene elements in corners (wheat stalks, sun rays)
3. Keep product lighting and composition exactly the same
4. Maintain professional e-commerce quality
5. Add brand watermark in bottom right corner
Style: Update should be seamless and natural-looking.
"""
edited_image = generator.edit_image(
image_path=reference_image,
prompt=prompt,
preserve_style=True
)
print(f"Edited image: {edited_image['file_path']}")
from nanobana_generator import NanoBananaGenerator
generator = NanoBananaGenerator()
# Create multi-turn conversation for step-by-step guide
conversation = generator.multi_turn_chat(
initial_prompt="Create a 5-panel comic guide explaining soil testing process for farmers",
iterations=5
)
# conversation contains all images and text history
for i, turn in enumerate(conversation['history']):
print(f"Step {i+1}: {turn['message']}")
if 'image_path' in turn:
print(f" Image: {turn['image_path']}")
from nanobana_generator import NanoBananaGenerator
generator = NanoBananaGenerator()
prompt = """
Visualize current weather forecast for next 5 days in Cedar Creek, TX as a clean, modern weather dashboard.
Include:
- Top section: 5-day forecast with daily high/low temperatures, weather icons, and precipitation chances
- Center panel: Temperature trend line graph showing warming trend from Monday to Friday
- Bottom panel: Farm action recommendations for each day (Monday: "Prepare freeze protection", Tuesday: "Monitor for ice melt")
Data accuracy: Use realistic temperatures based on current forecast (highs in 50s-60s F, lows in 20s-30s F). Include date labels and day abbreviations.
Style: Modern dashboard UI with data visualization, clean lines, and agricultural color scheme.
"""
image = generator.generate_image(
prompt=prompt,
use_search=True, # Enable Google Search for current weather data
resolution="2K",
aspect_ratio="16:9"
)
print(f"Generated weather dashboard: {image['file_path']}")
print(f"Grounding data: {image['grounding_metadata']}")
from nanobana_generator import NanoBananaGenerator
generator = NanoBananaGenerator()
prompt = """
Create a detailed technical cross-section diagram of a vermicomposting worm bin.
Visual elements to include:
- Cross-section view showing layers (top to bottom)
- Layer 1: Fresh food scraps and bedding (first 3 inches)
- Layer 2: Active composting zone with worms (middle 6 inches)
- Layer 3: Finished vermicast/castings ready for harvest (bottom 3 inches)
- Labels for each layer with dimensions
- Worm locations and movement arrows
- Drainage holes at bottom
- Air ventilation indicators
Style: Technical illustration style like agricultural extension diagrams. Clean lines, clear labels, cross-hatching for material differences. Include scale indicator (1 inch = 2.54 cm).
Title: "Vermicomposting Worm Bin Cross-Section"
Subtitle: "Layer Structure and Composting Zones"
"""
image = generator.generate_image(
prompt=prompt,
aspect_ratio="1:1",
resolution="2K"
)
print(f"Generated technical diagram: {image['file_path']}")
Problem: API key invalid or authentication fails
Solutions:
Problem: Generated images don't match expectations
Solutions:
Problem: API returns rate limit errors
Solutions:
Problem: Google Search grounding returns outdated or incorrect data
Solutions:
Problem: Complex multi-panel or detailed prompts fail to process
Solutions: