Nano Banana Use
- 70 installs
- 124 repo stars
- Updated February 6, 2026
- cnemri/google-genai-skills
Generates, edits, and composes images with Gemini Nano Banana models via portable uv-run Python scripts, with API-key or Vertex AI auth.
About
Runs image generation, editing, and multi-image composition through Gemini Nano Banana models via ready-made scripts. A developer uses it to create or modify images from the CLI without writing SDK code.
- generate, edit, and compose scripts run with uv
- Configurable model, aspect ratio, and safety filter
Nano Banana Use by the numbers
- 70 all-time installs (skills.sh)
- Ranked #848 of 1,335 Generative Media skills by installs in the Skillselion catalog
- Data as of Jul 29, 2026 (Skillselion catalog sync)
npx skills add https://github.com/cnemri/google-genai-skills --skill nano-banana-useAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 70 |
|---|---|
| repo stars | ★ 124 |
| Last updated | February 6, 2026 |
| Repository | cnemri/google-genai-skills ↗ |
What it does
Generates, edits, and composes images with Gemini Nano Banana models via portable uv-run Python scripts, with API-key or Vertex AI auth.
Files
Nano Banana Use
Use this skill to generate, edit, and compose images using Gemini's Nano Banana models (gemini-2.5-flash-image and gemini-3-pro-image-preview).
This skill uses portable Python scripts managed by uv.
Prerequisites
Ensure you have one of the following authentication methods configured in your environment:
1. API Key:
-
GOOGLE_API_KEYorGEMINI_API_KEY
2. Vertex AI:
-
GOOGLE_CLOUD_PROJECT -
GOOGLE_CLOUD_LOCATION -
GOOGLE_GENAI_USE_VERTEXAI=1
Usage
Generate an Image
Step 1: Confirm Parameters Before running the script, confirm the following parameters with the user or state the defaults you will use:
- Prompt: The image description.
- Model: Default is
gemini-3-pro-image-preview. - Aspect Ratio: Default is
1:1. - Safety Filter: Default is
BLOCK_MEDIUM_AND_ABOVE.
Step 2: Run the Script Run the python script using uv:
uv run skills/nano-banana-use/scripts/generate_image.py "A futuristic banana city" --output city.pngEdit an Image
Modify an existing image based on a text prompt.
uv run skills/nano-banana-use/scripts/edit_image.py original.png "Make the sky purple" --output edited.pngCompose Images
Generate a new image based on multiple input images and a prompt.
uv run skills/nano-banana-use/scripts/compose_image.py --image style.png --image subject.jpg "A painting of the subject in the style of the first image" --output composition.pngOptions
-
prompt: The text description of the image. -
--model: The model to use. Defaults togemini-3-pro-image-preview. -
--output: The filename for the saved image. Defaults togenerated_image.png. -
--aspect-ratio: The aspect ratio of the generated image. Defaults to1:1. Supported:1:1,16:9,4:3,3:4,9:16. -
--safety-filter-level: Safety filter threshold. Defaults toBLOCK_MEDIUM_AND_ABOVE.
# /// script
# requires-python = ">=3.11"
# dependencies = [
# "google-genai",
# "python-dotenv",
# "pillow",
# ]
# ///
import os
from dotenv import load_dotenv
import argparse
import sys
from google import genai
from google.genai import types
from PIL import Image
import io
load_dotenv()
def get_client():
api_key = os.environ.get("GOOGLE_API_KEY") or os.environ.get("GEMINI_API_KEY")
if api_key:
return genai.Client(api_key=api_key)
project = os.environ.get("GOOGLE_CLOUD_PROJECT")
location = os.environ.get("GOOGLE_CLOUD_LOCATION")
use_vertex = os.environ.get("GOOGLE_GENAI_USE_VERTEXAI", "").lower()
if project and location and use_vertex in ("1", "true"):
return genai.Client(vertexai=True, project=project, location=location)
print("=" * 60)
print("ERROR: Missing required environment variables!")
print("=" * 60)
print()
print("Please create a .env file in your project root or add the")
print("following environment variables to your existing .env file:")
print()
print("Option 1 - Using Gemini API Key:")
print(" GOOGLE_API_KEY=your-api-key-here")
print()
print("Option 2 - Using Vertex AI:")
print(" GOOGLE_CLOUD_PROJECT=your-project-id")
print(" GOOGLE_CLOUD_LOCATION=global")
print(" GOOGLE_GENAI_USE_VERTEXAI=1")
print()
print("Note: For Vertex AI, location must be 'global'.")
print("=" * 60)
sys.exit(1)
def main():
parser = argparse.ArgumentParser(description="Compose images using Nano Banana models.")
parser.add_argument("--image", action='append', required=True, help="Path to input image(s). Can be used multiple times.")
parser.add_argument("prompt", help="Instruction for composing the image")
parser.add_argument("--model", default="gemini-3-pro-image-preview", help="Model to use (default: gemini-3-pro-image-preview)")
parser.add_argument("--output", default="composed_image.png", help="Output filename")
parser.add_argument("--aspect-ratio", default="1:1", help="Aspect ratio (e.g., 1:1, 16:9)")
parser.add_argument("--safety-filter-level", default="BLOCK_NONE", help="Safety filter level")
args = parser.parse_args()
client = get_client()
try:
contents = []
for img_path in args.image:
if not os.path.exists(img_path):
print(f"Error: Input image '{img_path}' not found.")
sys.exit(1)
with open(img_path, "rb") as f:
image_bytes = f.read()
# Determine mime type roughly
mime_type = "image/jpeg"
if img_path.lower().endswith(".png"):
mime_type = "image/png"
elif img_path.lower().endswith(".webp"):
mime_type = "image/webp"
contents.append(types.Part.from_bytes(data=image_bytes, mime_type=mime_type))
contents.append(args.prompt)
response = client.models.generate_content(
model=args.model,
contents=contents,
config=types.GenerateContentConfig(
response_modalities=['IMAGE'],
image_config=types.ImageConfig(
aspect_ratio=args.aspect_ratio,
),
safety_settings=[
types.SafetySetting(
category="HARM_CATEGORY_SEXUALLY_EXPLICIT",
threshold=args.safety_filter_level
),
types.SafetySetting(
category="HARM_CATEGORY_DANGEROUS_CONTENT",
threshold=args.safety_filter_level
),
types.SafetySetting(
category="HARM_CATEGORY_HARASSMENT",
threshold=args.safety_filter_level
),
types.SafetySetting(
category="HARM_CATEGORY_HATE_SPEECH",
threshold=args.safety_filter_level
),
]
)
)
if response.candidates and response.candidates[0].content.parts:
for part in response.candidates[0].content.parts:
if part.inline_data:
img = Image.open(io.BytesIO(part.inline_data.data))
img.save(args.output)
print(f"Image saved to {args.output}")
return
print("No image generated.")
except Exception as e:
print(f"Error composing image: {e}")
sys.exit(1)
if __name__ == "__main__":
main()
# /// script
# requires-python = ">=3.11"
# dependencies = [
# "google-genai",
# "python-dotenv",
# "pillow",
# ]
# ///
import os
from dotenv import load_dotenv
import argparse
import sys
from google import genai
from google.genai import types
from PIL import Image
import io
load_dotenv()
def get_client():
api_key = os.environ.get("GOOGLE_API_KEY") or os.environ.get("GEMINI_API_KEY")
if api_key:
return genai.Client(api_key=api_key)
project = os.environ.get("GOOGLE_CLOUD_PROJECT")
location = os.environ.get("GOOGLE_CLOUD_LOCATION")
use_vertex = os.environ.get("GOOGLE_GENAI_USE_VERTEXAI", "").lower()
if project and location and use_vertex in ("1", "true"):
return genai.Client(vertexai=True, project=project, location=location)
print("=" * 60)
print("ERROR: Missing required environment variables!")
print("=" * 60)
print()
print("Please create a .env file in your project root or add the")
print("following environment variables to your existing .env file:")
print()
print("Option 1 - Using Gemini API Key:")
print(" GOOGLE_API_KEY=your-api-key-here")
print()
print("Option 2 - Using Vertex AI:")
print(" GOOGLE_CLOUD_PROJECT=your-project-id")
print(" GOOGLE_CLOUD_LOCATION=global")
print(" GOOGLE_GENAI_USE_VERTEXAI=1")
print()
print("Note: For Vertex AI, location must be 'global'.")
print("=" * 60)
sys.exit(1)
def main():
parser = argparse.ArgumentParser(description="Edit images using Nano Banana models.")
parser.add_argument("image", help="Path to the input image to edit")
parser.add_argument("prompt", help="Instruction for editing the image")
parser.add_argument("--model", default="gemini-3-pro-image-preview", help="Model to use (default: gemini-3-pro-image-preview)")
parser.add_argument("--output", default="edited_image.png", help="Output filename")
parser.add_argument("--aspect-ratio", default="1:1", help="Aspect ratio (e.g., 1:1, 16:9)")
parser.add_argument("--safety-filter-level", default="BLOCK_NONE", help="Safety filter level")
args = parser.parse_args()
if not os.path.exists(args.image):
print(f"Error: Input image '{args.image}' not found.")
sys.exit(1)
client = get_client()
try:
# Load the input image
with open(args.image, "rb") as f:
image_bytes = f.read()
# Determine mime type roughly
mime_type = "image/jpeg"
if args.image.lower().endswith(".png"):
mime_type = "image/png"
elif args.image.lower().endswith(".webp"):
mime_type = "image/webp"
response = client.models.generate_content(
model=args.model,
contents=[
types.Part.from_bytes(data=image_bytes, mime_type=mime_type),
args.prompt
],
config=types.GenerateContentConfig(
response_modalities=['IMAGE'],
image_config=types.ImageConfig(
aspect_ratio=args.aspect_ratio,
),
safety_settings=[
types.SafetySetting(
category="HARM_CATEGORY_SEXUALLY_EXPLICIT",
threshold=args.safety_filter_level
),
types.SafetySetting(
category="HARM_CATEGORY_DANGEROUS_CONTENT",
threshold=args.safety_filter_level
),
types.SafetySetting(
category="HARM_CATEGORY_HARASSMENT",
threshold=args.safety_filter_level
),
types.SafetySetting(
category="HARM_CATEGORY_HATE_SPEECH",
threshold=args.safety_filter_level
),
]
)
)
if response.candidates and response.candidates[0].content.parts:
for part in response.candidates[0].content.parts:
if part.inline_data:
img = Image.open(io.BytesIO(part.inline_data.data))
img.save(args.output)
print(f"Image saved to {args.output}")
return
print("No image generated.")
except Exception as e:
print(f"Error editing image: {e}")
sys.exit(1)
if __name__ == "__main__":
main()
# /// script
# requires-python = ">=3.11"
# dependencies = [
# "google-genai",
# "python-dotenv",
# "pillow",
# ]
# ///
import os
from dotenv import load_dotenv
import argparse
import sys
from google import genai
from google.genai import types
from PIL import Image
import io
load_dotenv()
def get_client():
api_key = os.environ.get("GOOGLE_API_KEY") or os.environ.get("GEMINI_API_KEY")
if api_key:
return genai.Client(api_key=api_key)
project = os.environ.get("GOOGLE_CLOUD_PROJECT")
location = os.environ.get("GOOGLE_CLOUD_LOCATION")
use_vertex = os.environ.get("GOOGLE_GENAI_USE_VERTEXAI", "").lower()
if project and location and use_vertex in ("1", "true"):
return genai.Client(vertexai=True, project=project, location=location)
print("=" * 60)
print("ERROR: Missing required environment variables!")
print("=" * 60)
print()
print("Please create a .env file in your project root or add the")
print("following environment variables to your existing .env file:")
print()
print("Option 1 - Using Gemini API Key:")
print(" GOOGLE_API_KEY=your-api-key-here")
print()
print("Option 2 - Using Vertex AI:")
print(" GOOGLE_CLOUD_PROJECT=your-project-id")
print(" GOOGLE_CLOUD_LOCATION=global")
print(" GOOGLE_GENAI_USE_VERTEXAI=1")
print()
print("Note: For Vertex AI, location must be 'global'.")
print("=" * 60)
sys.exit(1)
def main():
parser = argparse.ArgumentParser(description="Generate images using Nano Banana models.")
parser.add_argument("prompt", help="The text prompt for image generation")
parser.add_argument("--model", default="gemini-3-pro-image-preview", help="Model to use (default: gemini-3-pro-image-preview)")
parser.add_argument("--output", default="generated_image.png", help="Output filename")
parser.add_argument("--aspect-ratio", default="1:1", help="Aspect ratio (e.g., 1:1, 16:9, 4:3, 3:4, 9:16)")
parser.add_argument("--safety-filter-level", default="BLOCK_NONE", help="Safety filter level")
args = parser.parse_args()
client = get_client()
try:
response = client.models.generate_content(
model=args.model,
contents=args.prompt,
config=types.GenerateContentConfig(
response_modalities=['IMAGE'],
image_config=types.ImageConfig(
aspect_ratio=args.aspect_ratio,
),
safety_settings=[
types.SafetySetting(
category="HARM_CATEGORY_SEXUALLY_EXPLICIT",
threshold=args.safety_filter_level
),
types.SafetySetting(
category="HARM_CATEGORY_DANGEROUS_CONTENT",
threshold=args.safety_filter_level
),
types.SafetySetting(
category="HARM_CATEGORY_HARASSMENT",
threshold=args.safety_filter_level
),
types.SafetySetting(
category="HARM_CATEGORY_HATE_SPEECH",
threshold=args.safety_filter_level
),
]
)
)
if response.candidates and response.candidates[0].content.parts:
for part in response.candidates[0].content.parts:
if part.inline_data:
img = Image.open(io.BytesIO(part.inline_data.data))
img.save(args.output)
print(f"Image saved to {args.output}")
return
print("No image generated.")
except Exception as e:
print(f"Error generating image: {e}")
sys.exit(1)
if __name__ == "__main__":
main()
Related skills
Remotion Best PracticesGet Remotion-specific coding guidance that prevents common video rendering mistakes when creating animated React videos.451k4.2k
Ai Video GenerationGenerate videos using 40+ AI models including Veo, Seedance, and others via inference.sh442k665Ai Image GenerationGenerate images using 50+ AI models including FLUX, Gemini, and Replicate via inference.sh441k665Ai Avatar VideoGenerate videos featuring AI avatars for product demos, tutorials, marketing, or explainers441k665Remotion RenderRender videos programmatically using React components and Remotion framework441k665
Remotion RenderGenerate high-quality MP4 videos from React code using Remotion inside an AI coding agent.363k665