메뉴
Use when "nanobanana", "generate image", "create image", "edit image", "AI drawing", "Gemini image", "image generation"
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Use when evaluating agent performance, building test frameworks, measuring quality, or asking about "agent evaluation", "LLM-as-judge", "agent testing", "quality metrics", "evaluation rubrics", "agent benchmarks"
Use when "codex", "use gpt", "gpt-5", "openai codex", "let openai", "full-auto", "autonomous code generation"
Use when compressing agent context, implementing conversation summarization, reducing token usage in long sessions, or asking about "context compression", "conversation history", "token optimization", "context limits", "summarization strategies"
Use when diagnosing agent failures, debugging lost-in-middle issues, understanding context poisoning, or asking about "context degradation", "lost in middle", "context poisoning", "attention patterns", "context clash", "agent performance drops"
Use when optimizing agent context, reducing token costs, implementing KV-cache optimization, or asking about "context optimization", "token reduction", "context limits", "observation masking", "context budgeting", "context partitioning"
Use when "CrewAI", "multi-agent systems", "agent orchestration", "AI crews", or asking about "autonomous agents", "agent collaboration", "role-based agents", "agent workflows", "AI team coordination"
| name | nanobanana |
| description | Use when "nanobanana", "generate image", "create image", "edit image", "AI drawing", "Gemini image", "image generation" |
| version | 1.0.0 |
Generate and edit images using Google Gemini API.
| Requirement | Setup |
|---|---|
| API Key | Export GEMINI_API_KEY or add to ~/.nanobanana.env |
| Dependencies | pip install google-genai Pillow python-dotenv |
| Script | ${CLAUDE_PLUGIN_ROOT}/skills/core/nanobanana/nanobanana.py |
| Task | Command |
|---|---|
| Generate image | python3 nanobanana.py --prompt "description" --output "file.png" |
| Edit image | python3 nanobanana.py --prompt "changes" --input source.png --output "edited.png" |
| Size | Ratio | Use Case |
|---|---|---|
1024x1024 | 1:1 | Square, logos |
768x1344 | 9:16 | Portrait, stories (default) |
1344x768 | 16:9 | Landscape, wallpapers |
832x1248 | 2:3 | Portrait photos |
1248x832 | 3:2 | Landscape photos |
1536x672 | 21:9 | Ultra-wide |
Use --size WIDTHxHEIGHT to specify.
| Model | Trade-off |
|---|---|
gemini-3-pro-image-preview | Higher quality (default) |
gemini-2.5-flash-image | Faster generation |
Use --model MODEL to specify.
| Resolution | Use Case |
|---|---|
1K | Testing, drafts (default) |
2K | Good quality |
4K | Final output, print |
Use --resolution RES to specify.
| Practice | Why |
|---|---|
| Be descriptive | Include style, mood, colors, composition |
| Use 1:1 for logos | Clean square format |
| Use 9:16 for stories | Standard mobile format |
| Use 16:9 for wallpapers | Standard widescreen |
| Start with 1K | Test before using higher resolution |
| Use flash model for iteration | Save time during drafting |
Generate landscape:
python3 nanobanana.py --prompt "Mountain sunset with lake" --size 1344x768 --output "landscape.png"
Generate logo:
python3 nanobanana.py --prompt "Minimalist tech logo" --size 1024x1024 --output "logo.png"
Edit existing image:
python3 nanobanana.py --prompt "Add rainbow to sky" --input photo.png --output "edited.png"
High quality output:
python3 nanobanana.py --prompt "Professional portrait" --resolution 2K --output "portrait.png"
| Issue | Solution |
|---|---|
| Script fails | Check GEMINI_API_KEY is set |
| No image generated | Make prompt more specific |
| Can't read input | Verify file exists and is readable |
| Can't write output | Check directory is writable |