Compétences nano-banana-2
📦

nano-banana-2

Sûr

Gemini Flash로 이미지 생성

Également disponible depuis: inferen-sh,inference-sh-9,tul-sh,inference-shell

inference.sh CLI를 통해 Google Gemini 3.1 Flash Image 모델로 놀라운 이미지를 생성합니다. 텍스트에서 이미지 생성, 사진 편집, 여러 변형 توليد sowie Google Search 그라운딩을 통한 실시간 컨텍스트로 이미지를 생성할 수 있습니다.

Prend en charge: Claude Codex Code(CC)
📊 70 Adéquat
1

Télécharger le ZIP du skill

2

Importer dans Claude

Allez dans Paramètres → Capacités → Skills → Importer un skill

3

Activez et commencez à utiliser

Tester

Utilisation de "nano-banana-2". a banana in space, photorealistic

Résultat attendu:

Returns generated image URL(s) with metadata including description and pricing info

Utilisation de "nano-banana-2". Minimalist logo for coffee shop, 4 variations

Résultat attendu:

Returns array of 4 logo image URLs with descriptions

Audit de sécurité

Sûr
v1 • 3/8/2026

All static findings are false positives. The skill is a documentation wrapper for inference.sh CLI that provides usage examples for Google Gemini 3.1 Flash Image Preview. The detected backticks are markdown code block delimiters, not Ruby execution. URLs are legitimate service endpoints required for the image generation functionality. No cryptographic code or malicious patterns exist in this skill.

1
Fichiers analysés
167
Lignes analysées
0
résultats
1
Total des audits
Aucun problème de sécurité trouvé

Motifs détectés

External Command Detection (False Positive)Hardcoded URLs (False Positive)Weak Cryptographic Algorithm (False Positive)
Audité par: claude

Score de qualité

38
Architecture
100
Maintenabilité
85
Contenu
35
Communauté
100
Sécurité
83
Conformité aux spécifications

Ce que vous pouvez construire

빠른 이미지 생성

웹사이트 에셋, 소셜 미디어 또는 프레젠테이션을 위한 텍스트 프롬프트로 단일 이미지 생성

제품 시각화

전자상거래 또는 광고 캠페인을 위한 제품 이미지의 여러 변형 생성

개념 아트 생성

현재 이벤트 또는 트렌딩 주제를 위해 Google Search 그라운딩으로 예술적 시각화 생성

Essayez ces prompts

Basic Image Generation 
Generate an image of [describe your image] using nano-banana-2 skill
Multiple Variations 
Create 4 variations of [describe subject] with different styles using nano-banana-2
Image Editing 
Edit the uploaded image to [describe changes] using nano-banana-2
High-Resolution Production 
Generate a 4K resolution image of [describe scene] with [style details] using nano-banana-2

Bonnes pratiques

  • Be specific in prompts - include style, lighting, composition details for better results
  • Use aspect ratio matching your intended output (16:9 for landscape, 9:16 for social)
  • Enable Google Search grounding for current events or accurate location references

Éviter

  • Avoid vague prompts like 'make something cool' - be specific about subject and style
  • Do not exceed 14 input images for editing - will cause errors
  • Avoid requesting resolutions higher than 4K - not supported

Foire aux questions

What is nano-banana-2?
Nano-banana-2 is a skill for generating images using Google Gemini 3.1 Flash Image Preview through the inference.sh CLI tool.
Do I need an account to use this skill?
Yes, you need to install inference.sh CLI and create an account at inference.sh to use this skill.
Can I edit existing images?
Yes, you can provide up to 14 input images for editing tasks. The model will modify them based on your prompt.
What image formats are supported?
Input images can be JPEG, PNG, or WebP format. Output format is configurable.
What is Google Search grounding?
When enabled, the model uses real-time Google Search to provide accurate, current information in your generated images, useful for weather, news, or location-specific content.
What is the resolution limit?
Maximum resolution is 4K. You can also choose 1K or 2K resolutions.

Détails du développeur

Auteur

toolshell

Licence

MIT

Dépôt

https://github.com/toolshell/skills/tree/main/tools/image/nano-banana-2/

Réf

main

Structure de fichiers

📄 SKILL.md