nanobanana

Name: nanobanana
Author: ReScienceLab

2.5kinstalls115forks

How to install nanobanana?

resciencelab/opc-skills/nanobanana

$npx -y skills add ReScienceLab/opc-skills --skill nanobanana

Installs into the current project.

Summary

Generate and edit images using Google Gemini 3 Pro Image (Nano Banana Pro). Supports text-to-image, image editing, various aspect ratios, and high-resolution output (2K/4K). Use when user wants to generate images, create images, use Gemini image generation, or do AI image generat

Files · 1

View on GitHub

SKILL.md

1	# Nano Banana - AI Image Generation
2
3	Generate and edit images using Google's Gemini 3 Pro Image model (`gemini-3-pro-image-preview`, nicknamed "Nano Banana Pro" 🍌).
4
5	## Prerequisites
6
7	Required:
8	- `GEMINI_API_KEY` - Get from [Google AI Studio](https://aistudio.google.com/apikey)
9	- Python 3.10+ with `google-genai` package
10
11	Install dependencies:
12	```bash
13	pip install google-genai pillow
14	```
15
16	## Quick Start
17
18	### Generate an image:
19	```bash
20	python3 <skill_dir>/scripts/generate.py "a cute robot mascot, pixel art style" -o robot.png
21	```
22
23	### Edit an existing image:
24	```bash
25	python3 <skill_dir>/scripts/generate.py "make the background blue" -i input.jpg -o output.png
26	```
27
28	### Generate with specific aspect ratio:
29	```bash
30	python3 <skill_dir>/scripts/generate.py "cinematic landscape" --ratio 21:9 -o landscape.png
31	```
32
33	### Generate high-resolution 4K image:
34	```bash
35	python3 <skill_dir>/scripts/generate.py "professional product photo" --size 4K -o product.png
36	```
37
38	## Script Reference
39
40	### `scripts/generate.py`
41
42	Main image generation script.
43
44	```
45	Usage: generate.py [OPTIONS] PROMPT
46
47	Arguments:
48	PROMPT Text prompt for image generation
49
50	Options:
51	-o, --output PATH Output file path (default: auto-generated)
52	-i, --input PATH Input image for editing (optional)
53	-r, --ratio RATIO Aspect ratio (1:1, 16:9, 9:16, 21:9, etc.)
54	-s, --size SIZE Image size: 2K or 4K (default: standard)
55	--search Enable Google Search grounding for accuracy
56	-v, --verbose Show detailed output
57	```
58
59	Supported aspect ratios:
60	- `1:1` - Square (default)
61	- `2:3`, `3:2` - Portrait/Landscape
62	- `3:4`, `4:3` - Standard
63	- `4:5`, `5:4` - Photo
64	- `9:16`, `16:9` - Widescreen
65	- `21:9` - Ultra-wide/Cinematic
66
67	### `scripts/batch_generate.py`
68
69	Generate multiple images with sequential naming.
70
71	```
72	Usage: batch_generate.py [OPTIONS] PROMPT
73
74	Arguments:
75	PROMPT Text prompt for image generation
76
77	Options:
78	-n, --count N Number of images to generate (default: 10)
79	-d, --dir PATH Output directory
80	-p, --prefix STR Filename prefix (default: "image")
81	-r, --ratio RATIO Aspect ratio
82	-s, --size SIZE Image size (2K/4K)
83	--delay SECONDS Delay between generations (default: 3)
84	```
85
86	Example:
87	```bash
88	python3 <skill_dir>/scripts/batch_generate.py "pixel art logo" -n 20 -d ./logos -p logo
89	```
90
91	## Python API
92
93	You can also use the module directly:
94
95	```python
96	from generate import generate_image, edit_image
97
98	# Generate image
99	result = generate_image(
100	prompt="a futuristic city at night",
101	output_path="city.png",
102	aspect_ratio="16:9",
103	image_size="4K"
104	)
105
106	# Edit existing image
107	result = edit_image(
108	prompt="add flying cars to the sky",
109	input_path="city.png",
110	output_path="city_edited.png"
111	)
112	```
113
114	## Environment Variables
115
116	\| Variable \| Description \| Default \|
117	\|----------\|-------------\|---------\|
118	\| `GEMINI_API_KEY` \| Google Gemini API key \| Required \|
119	\| `IMAGE_OUTPUT_DIR` \| Default output directory \| `./nanobanana-images` \|
120
121	## Features
122
123	### Text-to-Image Generation
124	Create images from text descriptions. The model excels at:
125	- Photorealistic images
126	- Artistic styles (pixel art, illustration, etc.)
127	- Product photography
128	- Landscapes and scenes
129
130	### Image Editing
131	Transform existing images with natural language:
132	- Style transfer
133	- Object addition/removal
134	- Background changes
135	- Color adjustments
136
137	### High-Resolution Output
138	- Standard: Fast generation, good quality
139	- 2K: Enhanced detail (2048px)
140	- 4K: Maximum quality (3840px), best for text rendering
141
142	### Google Search Grounding
143	Enable `--search` for factually accurate images involving:
144	- Real people, places, landmarks
145	- Current events
146	- Specific products or brands
147
148	## Best Practices
149
150	### Prompt Writing
151
152	Good prompts include:
153	- Subject description
154	- Style/aesthetic
155	- Lighting and mood
156	- Composition details
157	- Color palette
158
159	Example:
160	```
161	"A cozy coffee shop interior, warm lighting, vintage aesthetic,
162	wooden furniture, plants on shelves, morning sunlight through windows,
163	soft focus background, 35mm film photography style"
164	```
165
166	### Batch Generation Tips
167
168	1. Generate 10-20 variations to explore options
169	2. Use consistent prompts for style coherence
170	3. Add 3-5 second delays to avoid rate limits
171	4. Review results and iterate on best candidates
172
173	## Rate Limits
174
175	- Gemini API has usage quotas
176	- Add delays between batch generations
177	- Check your quota at [Google AI Studio](https://aistudio.google.com/)
178
179	## Troubleshooting
180
181	"API key not found"
182	- Set `GEMINI_API_KEY` environment variable
183	- Or pass via `--api-key` option
184
185	"No image in response"
186	- Prompt may have triggered safety filters
187	- Try rephra

People who liked this also viewed

6 picks

Type

nanobanana

2.5kinstalls115forks

View on GitHub

How to install nanobanana?

resciencelab/opc-skills/nanobanana

$npx -y skills add ReScienceLab/opc-skills --skill nanobanana

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Nano Banana - AI Image Generation
2
3	Generate and edit images using Google's Gemini 3 Pro Image model (`gemini-3-pro-image-preview`, nicknamed "Nano Banana Pro" 🍌).
4
5	## Prerequisites
6
7	Required:
8	- `GEMINI_API_KEY` - Get from [Google AI Studio](https://aistudio.google.com/apikey)
9	- Python 3.10+ with `google-genai` package
10
11	Install dependencies:
12	```bash
13	pip install google-genai pillow
14	```
15
16	## Quick Start
17
18	### Generate an image:
19	```bash
20	python3 <skill_dir>/scripts/generate.py "a cute robot mascot, pixel art style" -o robot.png
21	```
22
23	### Edit an existing image:
24	```bash
25	python3 <skill_dir>/scripts/generate.py "make the background blue" -i input.jpg -o output.png
26	```
27
28	### Generate with specific aspect ratio:
29	```bash
30	python3 <skill_dir>/scripts/generate.py "cinematic landscape" --ratio 21:9 -o landscape.png
31	```
32
33	### Generate high-resolution 4K image:
34	```bash
35	python3 <skill_dir>/scripts/generate.py "professional product photo" --size 4K -o product.png
36	```
37
38	## Script Reference
39
40	### `scripts/generate.py`
41
42	Main image generation script.
43
44	```
45	Usage: generate.py [OPTIONS] PROMPT
46
47	Arguments:
48	PROMPT Text prompt for image generation
49
50	Options:
51	-o, --output PATH Output file path (default: auto-generated)
52	-i, --input PATH Input image for editing (optional)
53	-r, --ratio RATIO Aspect ratio (1:1, 16:9, 9:16, 21:9, etc.)
54	-s, --size SIZE Image size: 2K or 4K (default: standard)
55	--search Enable Google Search grounding for accuracy
56	-v, --verbose Show detailed output
57	```
58
59	Supported aspect ratios:
60	- `1:1` - Square (default)
61	- `2:3`, `3:2` - Portrait/Landscape
62	- `3:4`, `4:3` - Standard
63	- `4:5`, `5:4` - Photo
64	- `9:16`, `16:9` - Widescreen
65	- `21:9` - Ultra-wide/Cinematic
66
67	### `scripts/batch_generate.py`
68
69	Generate multiple images with sequential naming.
70
71	```
72	Usage: batch_generate.py [OPTIONS] PROMPT
73
74	Arguments:
75	PROMPT Text prompt for image generation
76
77	Options:
78	-n, --count N Number of images to generate (default: 10)
79	-d, --dir PATH Output directory
80	-p, --prefix STR Filename prefix (default: "image")
81	-r, --ratio RATIO Aspect ratio
82	-s, --size SIZE Image size (2K/4K)
83	--delay SECONDS Delay between generations (default: 3)
84	```
85
86	Example:
87	```bash
88	python3 <skill_dir>/scripts/batch_generate.py "pixel art logo" -n 20 -d ./logos -p logo
89	```
90
91	## Python API
92
93	You can also use the module directly:
94
95	```python
96	from generate import generate_image, edit_image
97
98	# Generate image
99	result = generate_image(
100	prompt="a futuristic city at night",
101	output_path="city.png",
102	aspect_ratio="16:9",
103	image_size="4K"
104	)
105
106	# Edit existing image
107	result = edit_image(
108	prompt="add flying cars to the sky",
109	input_path="city.png",
110	output_path="city_edited.png"
111	)
112	```
113
114	## Environment Variables
115
116	\| Variable \| Description \| Default \|
117	\|----------\|-------------\|---------\|
118	\| `GEMINI_API_KEY` \| Google Gemini API key \| Required \|
119	\| `IMAGE_OUTPUT_DIR` \| Default output directory \| `./nanobanana-images` \|
120
121	## Features
122
123	### Text-to-Image Generation
124	Create images from text descriptions. The model excels at:
125	- Photorealistic images
126	- Artistic styles (pixel art, illustration, etc.)
127	- Product photography
128	- Landscapes and scenes
129
130	### Image Editing
131	Transform existing images with natural language:
132	- Style transfer
133	- Object addition/removal
134	- Background changes
135	- Color adjustments
136
137	### High-Resolution Output
138	- Standard: Fast generation, good quality
139	- 2K: Enhanced detail (2048px)
140	- 4K: Maximum quality (3840px), best for text rendering
141
142	### Google Search Grounding
143	Enable `--search` for factually accurate images involving:
144	- Real people, places, landmarks
145	- Current events
146	- Specific products or brands
147
148	## Best Practices
149
150	### Prompt Writing
151
152	Good prompts include:
153	- Subject description
154	- Style/aesthetic
155	- Lighting and mood
156	- Composition details
157	- Color palette
158
159	Example:
160	```
161	"A cozy coffee shop interior, warm lighting, vintage aesthetic,
162	wooden furniture, plants on shelves, morning sunlight through windows,
163	soft focus background, 35mm film photography style"
164	```
165
166	### Batch Generation Tips
167
168	1. Generate 10-20 variations to explore options
169	2. Use consistent prompts for style coherence
170	3. Add 3-5 second delays to avoid rate limits
171	4. Review results and iterate on best candidates
172
173	## Rate Limits
174
175	- Gemini API has usage quotas
176	- Add delays between batch generations
177	- Check your quota at [Google AI Studio](https://aistudio.google.com/)
178
179	## Troubleshooting
180
181	"API key not found"
182	- Set `GEMINI_API_KEY` environment variable
183	- Or pass via `--api-key` option
184
185	"No image in response"
186	- Prompt may have triggered safety filters
187	- Try rephra

People who liked this also viewed

6 picks

Type