Manual setup required. The maintainer's config contains paths only you know - edit the placeholders below before adding it to Claude Code.
1
Prepare the server locally
Run this once before adding it to Claude Code.
git clone https://github.com/brandon695/nano-banana-mcp
cd nano-banana-mcpThen follow the repository README for any remaining dependency or build steps.
2
Register it in Claude Code
claude mcp add -e "GEMINI_API_KEY=${GEMINI_API_KEY}" nano-banana -- node /path/to/nano-banana-mcp-local/dist/index.jsReplace any placeholder paths in the command with the real path on your machine.
Required:
GEMINI_API_KEY+ 1 optionalEnvironment Variables
Set these before running Nano Banana MCP.
VariableDescriptionRequired
GEMINI_API_KEYYour Google AI Studio API keyYesNANO_BANANA_MODELDefault model for all requests (flash or pro)NoAvailable Tools (3)
Once configured, Nano Banana MCP gives your AI agent access to:
generate_imageGenerates an image based on a text prompt with optional quality and aspect ratio settings.promptqualityaspectRatioedit_imageEdits an existing image based on a text prompt.promptqualitycontinue_editingContinues an existing image editing session.promptqualityTry It Out
After setup, try these prompts with your AI agent:
→Generate a dramatic photo of a futuristic city skyline at sunset.
→Create a 16:9 image of a cozy cabin in the woods during winter.
→Edit the previous image to add a golden retriever sitting on the porch.
→Generate a high-quality portrait of a robot reading a book using the pro model.
Prerequisites & system requirements
- An MCP-compatible client (Claude Code, Cursor, Windsurf, Claude Desktop, or Codex)
- Node.js 18+ with npm/npx installed
GEMINI_API_KEY— Your Google AI Studio API key
Keep this setup from going cold
Save the docs, env vars, and workflow around Nano Banana MCP in Conare so Claude Code, Codex, and Cursor remember it next time.