nano-banana-image-generator
Skill109 starsChanged 40 days agoMITView on GitHub
Generate images using Google's Nano Banana (Gemini 2.5 Flash Image). Use when creating infographics, diagrams, thumbnails, social media graphics, or any AI-generated images. Costs $0.039/image.
- Reads credentials
In this repository61 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- adviseSkill
- ai-crystallizeSkill
- analyze-kbSkill
- auto-discoverySkill
- benchmark-memorySkill
- brain-mergeSkill
- changelog-protocolSkill
- coherence-sweepSkill
- compute-lifecycleSkill
- create-articleSkill
- create-explanatory-imageSkill
- decideSkill
- deep-researchSkill
- detect-tensionsSkill
- dialecticSkill
- domain-watchSkill
- elicitation-techniquesSkill
- epistemic-classificationSkill
- epub-chapter-extractorSkill
- extract-document-insightsSkill
- extract-insightsSkill
- find-connectionsSkill
- first-runSkill
- get-perspective-onSkill
- get-youtube-transcriptSkill
- git-commit-pushSkill
- graduate-insightsSkill
- incubation-loopSkill
- ingest-sourceSkill
- insight-capture-formatSkill
- insight-interviewSkill
- integrate-recent-notesSkill
- learn-new-thingsSkill
- manage-reference-dataSkill
- manage-thinking-topicsSkill
- manage-watching-domainsSkill
- multi-format-book-extractorSkill
- propagate-changeSkill
- quick-searchSkill
- reasoning-checksSkill
- recallSkill
- ref-auditSkill
- ref-bridgeSkill
- ref-ingestSkill
- ref-querySkill
- ref-reconcileSkill
- ref-refreshSkill
- refresh-indexSkill
- ref-supersedeSkill
- scheduled-runSkill
- scope-mountSkill
- search-vaultSkill
- self-diagnosticSkill
- start-hereSkill
- synthesize-insightsSkill
- talkSkill
- test-memory-systemSkill
- think-about-itSkill
- update-changelogSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 69
DO:: Write narrative descriptions
- Distinctive: line 75
DON'T: Use keyword lists
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Nano Banana Image Generator
Generate images using Google's Gemini 2.5 Flash Image model (codename: Nano Banana).
State Dependencies
| Source | Location | Read | Write | Description |
| Generate Script | ~/.claude/skills/nano-banana-image-generator/scripts/generate.sh | ✓ | Main generation script | |
| Thumbnail Script | ~/.claude/skills/nano-banana-image-generator/scripts/generate_thumbnail.py | ✓ | 16:9 thumbnail generator | |
| Image Script | ~/.claude/skills/nano-banana-image-generator/scripts/generate_image.py | ✓ | Python with aspect ratio control | |
| Best Practices | ~/.claude/skills/nano-banana-image-generator/best_practices.md | ✓ | Prompt engineering guidelines | |
| API Key | $GOOGLE_API_KEY or $GEMINI_API_KEY | ✓ | Required for API access | |
| Output Directory | $OUTPUT_DIR or current directory | ✓ | Where images are saved | |
| Generated Image | User-specified path | ✓ | Output PNG file |
Quick Start
Generate a simple image:
bash ~/.claude/skills/nano-banana-image-generator/scripts/generate.sh "A red apple on white background" apple.pngGenerate a 16:9 thumbnail:
python3 ~/.claude/skills/nano-banana-image-generator/scripts/generate_thumbnail.py "YouTube thumbnail showing AI agents" /tmp/thumb.pngGenerate with custom output directory:
OUTPUT_DIR=/tmp bash ~/.claude/skills/nano-banana-image-generator/scripts/generate.sh "Sunset over mountains" sunset.pngScripts
| Script | Purpose | Output |
scripts/generate.sh | General image generation | 1024x1024 PNG |
scripts/generate_thumbnail.py | 16:9 thumbnails | 1344x768 PNG |
scripts/generate_image.py | Python with aspect ratio control | Configurable |
Pricing & Limits
- Cost: $0.039 per image
- Free tier: 500 requests/day
- Generation time: ~22 seconds
- Resolution: 1024x1024 (square), 1344x768 (16:9)
Best Practices
For detailed prompt engineering and brand styling guidelines, see best_practices.md.
Critical Constraints
| Element | Maximum |
| Visual boxes/shapes | 10 maximum |
| Text pieces/labels | 10 maximum |
| Hierarchy levels | 3 maximum |
Prompt Structure
DO: Write narrative descriptions
Create a simple infographic with black background showing a central concept
with 4-5 supporting elements. Use white text, DM Sans font, bold titles.DON'T: Use keyword lists
LinkedIn post, gradient, red, white, large text, 15 boxes, arrows everywhereBrand Styling (the user's)
Include in prompts for brand consistency:
Use the user's brand style: black background (#000000), white text (#ffffff),
DM Sans font, bold 700 weight for titles, clean minimal design.Use Cases
Perfect For:
- Social media infographics and carousels
- YouTube thumbnails
- Diagrams and technical documentation
- Product mockups
- Mobile-optimized content
Not Ideal For:
- Hyper-realistic photography
- Complex multi-font typography
- Fine detail technical renders
Environment Variables
GOOGLE_API_KEYorGEMINI_API_KEY- API key (required)OUTPUT_DIR- Output directory (default: current directory)
Examples
Infographic:
bash scripts/generate.sh "Simple framework diagram with black background. CENTER: 'AI Agents' in large white text. SURROUNDING: 4 icons for Planning, Memory, Tools, Learning. DM Sans font, minimal design." agents_framework.pngThumbnail:
python3 scripts/generate_thumbnail.py "Professional YouTube thumbnail: bold white text 'DEEP AGENTS' on dramatic black background with subtle blue glow, tech aesthetic" deep_agents_thumb.pngCompletion Checklist
- API key verified (GOOGLE_API_KEY or GEMINI_API_KEY)
- Prompt crafted following best practices (narrative description, not keyword lists)
- Critical constraints respected (max 10 boxes, max 10 text pieces, max 3 hierarchy levels)
- Appropriate script selected (generate.sh, generate_thumbnail.py, or generate_image.py)
- Output directory confirmed or created
- Image generated successfully (~22 seconds)
- Output file path provided to user
- Cost noted ($0.039 per image)
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

