Nano Banana 2 Skill

Nano Banana 2 — AI Image Generation Skill for OpenClaw

---

What Is This?

An OpenClaw skill powered by EvoLink. Install one skill, and your AI agent gets access to Google's Gemini 3.1 Flash image model — fast text-to-image and image editing through a single API.

Focused view of evolink-image. Install the full skill for 19 image models, video, and music.

🚀 Try All Models Now →

---

Installation

Quick Install (Recommended)

openclaw skills add https://github.com/EvoLinkAI/nano-banana-2-skill

Done. The skill is ready for your agent.

Manual Install

git clone https://github.com/EvoLinkAI/nano-banana-2-skill.git
cd nano-banana-2-skill
openclaw skills add .

---

Get API Key

1. Sign up at evolink.ai
2. Go to Dashboard → API Keys
3. Create a new Key
4. Set the environment variable:

export EVOLINK_API_KEY=your_key_here

Or just tell your OpenClaw agent: "Set my EvoLink API key to ..." — it'll handle the rest.

---

Image Generation

Generate and edit AI images through natural conversation with your OpenClaw agent.

Capabilities

• Text-to-Image — Describe what you want, get an image
• Image Editing — Transform existing images with reference
• Multi-Resolution — 1024×1024, 1024×1536, 1536×1024, and custom aspect ratios
• Batch Generation — Generate 1–4 variations in one call
• Aspect Ratios — 1:1, 16:9, 9:16, 4:3, 3:4, 2:3, 3:2, and more

Why Nano Banana 2?

• Google's latest — Built on Gemini 3.1 Flash, the newest generation architecture
• Fast generation — Optimized for speed without sacrificing quality
• Versatile — General-purpose image creation for any creative need
• Strong prompt adherence — Excellent at following complex, detailed descriptions

Lite Variant

nano-banana-2-lite [BETA] — Lightweight version for ultra-fast iterations when speed is the top priority.

Usage Examples

Just talk to your agent:

"Generate a watercolor painting of a mountain sunset"

"Edit this photo to make it look like a vintage film still"

"Create 4 variations of a logo design for a coffee shop"

"Make this image look more cinematic with warm tones"

The agent will guide you through missing details and handle generation.

Alternative Models

MCP Tools

Parameters

---

Setup — MCP Server Bridge

For the full tool experience, bridge the MCP server @evolinkai/evolink-media (npm):

Via mcporter:

mcporter call --stdio "npx -y @evolinkai/evolink-media@latest" list_models

Or add to mcporter config:

{
  "evolink-media": {
    "transport": "stdio",
    "command": "npx",
    "args": ["-y", "@evolinkai/evolink-media@latest"],
    "env": { "EVOLINK_API_KEY": "your-key-here" }
  }
}

Direct install (Claude Code):

claude mcp add evolink-media -e EVOLINK_API_KEY=your-key -- npx -y @evolinkai/evolink-media@latest

---

File Upload

For image editing or reference workflows, upload images first:

1. Call upload_file with file_path, base64_data, or file_url → get file_url (synchronous)
2. Use that file_url as image_urls for generate_image

Supported: Images (JPEG, PNG, GIF, WebP). Max 100MB. Files expire after 72 hours. Quota: 100 files (default) / 500 (VIP).

---

Workflow

1. Upload reference images if needed (via upload_file)
2. Call generate_image → get task_id
3. Poll check_task every 3–5s until completed
4. Download result URLs (expire in 24h)

---

Script Reference

The skill includes scripts/evolink-image-gen.sh for direct command-line usage:

# Text-to-image
./scripts/evolink-image-gen.sh "Watercolor mountain sunset" --size 1024x1536

# With reference image
./scripts/evolink-image-gen.sh "Make it look vintage" --image "https://example.com/photo.jpg"

# Batch generation (4 variations)
./scripts/evolink-image-gen.sh "Coffee shop logo design" --n 4

# Quick iteration with lite variant
./scripts/evolink-image-gen.sh "Abstract geometric pattern" --model nano-banana-2-lite

API Reference

Full API documentation: references/api-params.md

---

File Structure

.
├── README.md                    # English
├── README.zh-CN.md              # 简体中文
├── SKILL.md                     # OpenClaw skill definition
├── _meta.json                   # Skill metadata
├── references/
│   └── api-params.md            # Full API parameter docs
└── scripts/
    └── evolink-image-gen.sh     # Image generation script

---

Troubleshooting