# Gemini Nanobanana (Image Generation)

> Use this tool when you need to generate, edit, or compose images programmatically, and want to leverage AI-powered capabilities for tasks like automated content creation, design prototyping, or batch image processing. It takes in text prompts, image files, and configuration options as inputs, and outputs generated or edited images in various formats. Ideal for developers and content creators seeking advanced image generation capabilities with robust error handling and security measures.

Canonical page: https://skillsregistry.net/skills/junhan2-gemini-nanobanana  
JSON: https://api.skillsregistry.net/v1/skills/junhan2-gemini-nanobanana

## Description

A TypeScript MCP server that provides AI-powered image generation, editing, and composition capabilities through Google's Gemini 2.5 Flash Image preview API. Built by Junhan2 with comprehensive error handling, retry logic, and structured logging, it offers four core tools: text-to-image generation, single image editing with prompts, multi-image composition, and style transfer between images. The implementation features robust input validation with support for multiple image formats (PNG, JPEG, WebP, GIF), automatic file saving with collision handling, configurable transport modes (stdio/HTTP), and security measures including path traversal protection and request size limits. Designed for developers and content creators who need programmatic access to advanced image generation capabilities for workflows like automated content creation, design prototyping, or batch image processing tasks.

## Trust

- **Trust score (0–1):** 0.91
- **Verification tier:** verified
- **Last scanned:** 2026-09-28

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** mcp-remote
- **Runtime environment:** api
- **Category:** media
- **Updated:** 2026-09-28

## Source

- **Source listing:** [PulseMCP](https://www.pulsemcp.com/servers/junhan2-gemini-nanobanana)
- **Repository:** <https://github.com/junhan2/gemini-nanobanana-mcp>

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "junhan2-gemini-nanobanana"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/junhan2-gemini-nanobanana` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/junhan2-gemini-nanobanana/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt