NanoBananaMCP

A Model Context Protocol (MCP) server for AI image generation and editing using Google's Nano Banana model through the AceDataCloud API.

Generate and edit AI images directly from Claude, VS Code, or any MCP-compatible client.

Features

• Image Generation - Create high-quality images from text prompts
• Image Editing - Modify existing images or combine multiple images
• Virtual Try-On - Put clothing on people in photos
• Product Placement - Place products in realistic scenes
• Task Tracking - Monitor generation progress and retrieve results

Tool Reference

Connect: hosted OAuth, API token, or local stdio

The hosted endpoint is https://nanobanana.mcp.acedata.cloud/mcp. Choose one route for the MCP client:

The hosted service advertises OAuth metadata and Dynamic Client Registration (DCR). DCR registers the client application; it is not an API key. OAuth signs you in and the client sends the resulting Bearer token; it may reuse or create an API credential for the account. Browser sign-in still requires an AceDataCloud account. The hosted service can be metered: review current service documentation and displayed pricing before a real operation. Do not configure both an OAuth login and a fixed Authorization header for the same server.

Hosted OAuth examples

• Claude and Claude Desktop chat: Add a remote custom connector in Customize → Connectors → Add custom connector, enter https://nanobanana.mcp.acedata.cloud/mcp, select sign-in, and choose Register automatically if Claude asks how to register its OAuth client. Complete consent. Claude Desktop's local claude_desktop_config.json is a separate setup. Claude connector guide.
• Claude Code: claude mcp add --transport http --scope user nanobanana https://nanobanana.mcp.acedata.cloud/mcp, then claude mcp login nanobanana. Check /mcp. Claude Code MCP guide.
• Cursor: Add a remote server with only https://nanobanana.mcp.acedata.cloud/mcp. For a project, merge the entry below into <project>/.cursor/mcp.json; for personal use, use ~/.cursor/mcp.json. Cursor MCP guide.
• VS Code / Copilot: Run MCP: Add Server, select HTTP, enter https://nanobanana.mcp.acedata.cloud/mcp, then finish the browser sign-in. New portable workspace configs use <project>/.mcp.json; the VS Code-specific format below uses <project>/.vscode/mcp.json or the user profile. Check MCP: List Servers. VS Code MCP setup.
• Codex: codex mcp add nanobanana --url https://nanobanana.mcp.acedata.cloud/mcp, then codex mcp login nanobanana. Its user settings are in ~/.codex/config.toml. Official Codex MCP guide.

Cursor project config (OAuth):

{
  "mcpServers": {
    "nanobanana": {"url": "https://nanobanana.mcp.acedata.cloud/mcp"}
  }
}

VS Code-specific workspace config (OAuth):

{
  "servers": {
    "nanobanana": {"type": "http", "url": "https://nanobanana.mcp.acedata.cloud/mcp"}
  }
}

Hosted API token

Sign in at AceDataCloud Platform, open the service page, and obtain an API credential. A fixed Bearer header is useful when your client lacks OAuth; an invalid header does not fall back to OAuth in Claude Code. The header value is sensitive, so keep it out of committed files and screenshots.

For Claude Code, the shell expands the token when you add the server; treat the saved user MCP config as a secret:

export ACEDATACLOUD_API_TOKEN='YOUR_API_TOKEN'
claude mcp add --transport http --scope user nanobanana https://nanobanana.mcp.acedata.cloud/mcp \
  --header "Authorization: Bearer $ACEDATACLOUD_API_TOKEN"

For a Claude Code project config, put a variable reference in <project>/.mcp.json and set that variable in the environment that launches Claude Code:

{
  "mcpServers": {
    "nanobanana": {
      "type": "http",
      "url": "https://nanobanana.mcp.acedata.cloud/mcp",
      "headers": {"Authorization": "Bearer ${ACEDATACLOUD_API_TOKEN}"}
    }
  }
}

Cursor uses a different environment-variable syntax in ~/.cursor/mcp.json or an uncommitted project config:

{
  "mcpServers": {
    "nanobanana": {
      "url": "https://nanobanana.mcp.acedata.cloud/mcp",
      "headers": {"Authorization": "Bearer ${env:ACEDATACLOUD_API_TOKEN}"}
    }
  }
}

In VS Code, run MCP: Open User Configuration and merge this server plus its masked input; ${input:...} is for VS Code's user/workspace format and is not portable to the Agent Host .mcp.json format:

{
  "inputs": [
    {"id": "acedata-nanobanana-token", "type": "promptString", "description": "AceDataCloud API token", "password": true}
  ],
  "servers": {
    "nanobanana": {
      "type": "http",
      "url": "https://nanobanana.mcp.acedata.cloud/mcp",
      "headers": {"Authorization": "Bearer ${input:acedata-nanobanana-token}"}
    }
  }
}

For Cline, use its MCP configuration UI or CLI file ~/.cline/data/settings/cline_mcp_settings.json; its remote transport value is streamableHttp. For JetBrains AI Assistant, add a remote URL from Settings → Tools → AI Assistant → Model Context Protocol (MCP). For Zed, use a context_servers entry with the URL only for OAuth or add a local Bearer header. These clients have different configuration schemas; follow their current UI rather than copying another client's JSON. Cline · JetBrains · Zed.

Local stdio

Install the package and give the local process an API token:

python -m pip install mcp-nanobanana-pro
export ACEDATACLOUD_API_TOKEN='YOUR_API_TOKEN'
mcp-nanobanana-pro

For Claude Desktop local MCP, merge this entry into the file opened by its developer settings (~/Library/Application Support/Claude/claude_desktop_config.json on macOS). uvx requires uv on PATH:

{
  "mcpServers": {
    "nanobanana": {
      "command": "uvx",
      "args": ["mcp-nanobanana-pro"],
      "env": {"ACEDATACLOUD_API_TOKEN": "YOUR_API_TOKEN"}
    }
  }
}

Keep this user-level file private. Self-hosted HTTP uses mcp-nanobanana-pro --transport http --port 8000; expose it only with suitable network and TLS controls. Local execution still calls the AceDataCloud API.

Check before using the service

1. https://nanobanana.mcp.acedata.cloud/health returning {"status":"ok"} checks endpoint reachability only.
2. Confirm that the MCP client loads tools. The tool list shows MCP discovery, not downstream API access or balance.
3. If you need a full API check, call nanobanana_generate_image with your own valid input after reviewing current service documentation and displayed pricing. If the result contains a task ID, call nanobanana_get_task on that same ID until terminal success or failure. Do not resubmit the operation just to check progress.

For 401, check which auth route the client used and whether the token or OAuth session is valid. A 403 may mean an account permission or content moderation failure; read the returned error. Insufficient balance and downstream service failures need their own diagnosis. A listed tool or submitted task does not prove a successful result.

Available Tools

Image Generation

Tasks

Supported Models

All five models support generation and editing. The existing nano-banana, nano-banana-2-lite, nano-banana-2, and nano-banana-pro models also have :official variants. Nano Banana 2.1 has no :official variant; use exactly nano-banana-2.1. Omitting model still selects nano-banana.

Usage Examples

Nano Banana 2.1 Generation and Editing

Pass these arguments to nanobanana_generate_image:

{
  "prompt": "A blue ceramic vase on a cream background, soft side lighting, no text",
  "model": "nano-banana-2.1",
  "aspect_ratio": "1:1",
  "resolution": "2K",
  "count": 2
}

For nanobanana_edit_image, supply your reference image URL:

{
  "prompt": "Change the vase to green while preserving the background and lighting",
  "image_urls": ["https://example.com/vase.png"],
  "model": "nano-banana-2.1",
  "resolution": "4K"
}

Generation and editing submit asynchronously. Keep the returned task_id and query nanobanana_get_task until terminal success or failure; submission alone is not a completed image. Editing without aspect_ratio preserves the first reference image's aspect ratio; resolution defaults to 1K if omitted. Review the final image and pixel dimensions before using it.

Generate Image from Prompt

User: Create an image of a sunset over mountains

Claude: I'll generate that image for you.
[Calls nanobanana_generate_image with detailed prompt]

Virtual Try-On

User: Put this shirt on this model
[Provides two image URLs]

Claude: I'll combine these images.
[Calls nanobanana_edit_image with both image URLs]

Product Photography

User: Place this product in a modern kitchen scene
[Provides product image URL]

Claude: I'll create a product scene for you.
[Calls nanobanana_edit_image with scene description]

Prompt Writing Tips

For best results, include these elements in your prompts:

• Main Subject: What is the primary focus?
• Atmosphere: What mood should the image convey?
• Lighting: How is the scene illuminated?
• Camera/Lens: What photographic style? (85mm portrait, wide-angle, etc.)
• Quality Keywords: Technical descriptors (bokeh, film grain, HDR, etc.)

Example Prompt

A photorealistic close-up portrait of an elderly Japanese ceramicist
with deep wrinkles and a warm smile. Soft golden hour light streaming
through a window. Captured with an 85mm portrait lens, soft bokeh
background. Serene and masterful mood.

Configuration

Environment Variables

Command Line Options

mcp-nanobanana-pro --help

Options:
  --version          Show version
  --transport        Transport mode: stdio (default) or http
  --port             Port for HTTP transport (default: 8000)

Development

Setup Development Environment

# Clone repository
git clone https://github.com/AceDataCloud/NanoBananaMCP.git
cd NanoBananaMCP

# Create virtual environment
python -m venv .venv
source .venv/bin/activate  # or `.venv\Scripts\activate` on Windows

# Install with dev dependencies
pip install -e ".[dev,test]"

Run Tests

# Run unit tests
pytest

# Run with coverage
pytest --cov=core --cov=tools

# Run integration tests (requires API token)
pytest tests/test_integration.py -m integration

Code Quality

# Format code
ruff format .

# Lint code
ruff check .

# Type check
mypy core tools

Build & Publish

# Install build dependencies
pip install -e ".[release]"

# Build package
python -m build

# Upload to PyPI
twine upload dist/*

Project Structure

NanoBanana/
├── core/                   # Core modules
│   ├── __init__.py
│   ├── client.py          # HTTP client for NanoBanana API
│   ├── config.py          # Configuration management
│   ├── exceptions.py      # Custom exceptions
│   ├── server.py          # MCP server initialization
│   ├── types.py           # Type definitions
│   └── utils.py           # Utility functions
├── tools/                  # MCP tool definitions
│   ├── __init__.py
│   ├── image_tools.py     # Image generation/editing tools
│   └── task_tools.py      # Task query tools
├── prompts/                # MCP prompt templates
│   └── __init__.py
├── tests/                  # Test suite
├── deploy/                 # Deployment configs
│   └── production/
│       ├── deployment.yaml
│       ├── ingress.yaml
│       └── service.yaml
├── .env.example           # Environment template
├── .gitignore
├── Dockerfile             # Docker image for HTTP mode
├── docker-compose.yaml    # Docker Compose config
├── LICENSE
├── main.py                # Entry point
├── pyproject.toml         # Project configuration
└── README.md

API Reference

This server wraps the AceDataCloud NanoBanana API:

• NanoBanana Images API - Image generation and editing
• NanoBanana Tasks API - Task queries

Use Cases

• Portrait Enhancement - Try different clothing on the same person
• Product Scene Composition - Place white-background products in realistic environments
• Attribute Replacement - Change materials, colors, or variants
• Poster Quick Editing - Rapidly change styles or themes
• 2D to 3D Conversion - Convert images to 3D product mockups
• Image Restoration - Restore old or damaged photos

Contributing

Contributions are welcome! Please:

1. Fork the repository
2. Create a feature branch (git checkout -b feature/amazing)
3. Commit your changes (git commit -m 'Add amazing feature')
4. Push to the branch (git push origin feature/amazing)
5. Open a Pull Request

Documentation

Documentation

License

MIT License - see LICENSE for details.

Links

• AceDataCloud Platform
• Model Context Protocol
• MCP Python SDK

Made with love by AceDataCloud