Skip to main content
Glama
naporin0624

gpt-image-mcp

by naporin0624
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript

gpt-image MCP

npm version License: MIT OpenAI Documentation

MCP server for AI-powered image generation using OpenAI's gpt-image-1 and gpt-image-2 models with advanced text rendering and native transparency support.

Features

  • Advanced text rendering with gpt-image-1 - Crisp, legible typography and logos in generated images

  • Native transparency support - Built-in transparent background without post-processing

  • Multi-format output (PNG, JPEG, WebP) - Flexible format options with optimized compression

  • Flexible dimensions and aspect ratios - Square (1024×1024), landscape (1536×1024), and portrait (1024×1536)

  • Batch image editing capabilities - Process multiple images with parallel processing

  • Token-optimized MCP responses - Efficient response formats for MCP protocol limits

Related MCP server: GPT-Image-1 MCP Server

Installation

Recommended: Using npx

{
  "mcpServers": {
    "gpt-image-mcp": {
      "command": "npx",
      "args": ["@napolab/gpt-image-mcp"],
      "env": {
        "OPENAI_API_KEY": "sk-your-api-key"
      }
    }
  }
}

Alternative: Local Installation

npm install -g @napolab/gpt-image-mcp

Claude Desktop Configuration

Configure in ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):

{
  "mcpServers": {
    "gpt-image-mcp": {
      "command": "npx",
      "args": ["@napolab/gpt-image-mcp"],
      "env": {
        "OPENAI_API_KEY": "sk-your-api-key"
      }
    }
  }
}

Configuration

Environment Variables

Variable

Required

Default

Description

OPENAI_API_KEY

Yes

-

Your OpenAI API key

DEFAULT_OUTPUT_DIR

No

./generated_images

Default output directory

DEFAULT_IMAGE_SIZE

No

1024x1024

Default image dimensions

DEFAULT_IMAGE_QUALITY

No

standard

Default quality (standard/hd)

DEFAULT_OUTPUT_FORMAT

No

png

Default format (png/jpeg/webp)

Available Tools

generate-image

Generate images using gpt-image-1 with advanced text rendering and superior instruction following.

Parameters:

Parameter

Type

Required

Default

Description

prompt

string

Yes

-

Image description (English only)

aspect_ratio

string

No

square

"square", "landscape", or "portrait"

quality

string

No

standard

"standard" or "hd"

output_directory

string

No

./generated_images

Directory to save the image

filename

string

No

-

Custom filename

save_to_file

boolean

No

true

Whether to save locally

include_base64

boolean

No

false

Include base64 in response

Example:

await client.callTool("generate-image", {
  prompt: "A serene Japanese garden with cherry blossoms",
  aspect_ratio: "landscape",
  quality: "hd",
});

edit-image

Edit existing images with AI-powered modifications including inpainting, outpainting, style transfer, and background changes.

Parameters:

Parameter

Type

Required

Default

Description

source_image

object

Yes

-

Image input (URL, base64, or local file)

edit_prompt

string

Yes

-

Description of desired changes (English only)

edit_type

string

Yes

-

Type of edit to perform

strength

number

No

0.8

Edit strength (0.0 to 1.0)

preserve_composition

boolean

No

true

Maintain original composition

output_format

string

No

png

Output format

Edit Types:

  • inpaint - Fill in or modify specific areas

  • outpaint - Extend image beyond boundaries

  • background_change - Replace or modify background

  • style_transfer - Apply artistic styles

  • object_removal - Remove unwanted objects

  • variation - Create variations of original

Example:

await client.callTool("edit-image", {
  source_image: {
    type: "local",
    value: "/path/to/image.jpg",
  },
  edit_prompt: "Add a sunset sky background",
  edit_type: "background_change",
});

batch-edit

Apply the same edit to multiple images efficiently with parallel processing.

Parameters:

Parameter

Type

Required

Default

Description

images

array

Yes

-

Array of image inputs

edit_prompt

string

Yes

-

Edit description (English only)

edit_type

string

Yes

-

Type of edit to apply

batch_settings

object

No

-

Batch processing configuration

Example:

await client.callTool("batch-edit", {
  images: [
    { type: "local", value: "/path/to/image1.jpg" },
    { type: "local", value: "/path/to/image2.jpg" },
  ],
  edit_prompt: "Apply vintage sepia filter",
  edit_type: "style_transfer",
});

Usage Examples

Basic Image Generation

// Generate a simple image
const result = await client.callTool("generate-image", {
  prompt: "A modern minimalist logo design",
  aspect_ratio: "square",
  quality: "hd",
});

console.log("Generated image:", result.data.file_path);

Advanced Options

// Generate with all parameters
const result = await client.callTool("generate-image", {
  prompt: "Professional product photography of a smartphone",
  aspect_ratio: "portrait",
  quality: "hd",
  output_directory: "./product_images",
  filename: "smartphone_hero",
  output_format: "png",
  include_base64: true,
});

Image Editing

// Generate base image
const baseImage = await client.callTool("generate-image", {
  prompt: "A mountain landscape",
  aspect_ratio: "landscape",
});

// Edit the generated image
const editedImage = await client.callTool("edit-image", {
  source_image: {
    type: "local",
    value: baseImage.data.file_path,
  },
  edit_prompt: "Add dramatic storm clouds",
  edit_type: "background_change",
  strength: 0.7,
});

Batch Processing

// Process multiple images
const result = await client.callTool("batch-edit", {
  images: [
    { type: "local", value: "image1.jpg" },
    { type: "local", value: "image2.jpg" },
    { type: "local", value: "image3.jpg" },
  ],
  edit_prompt: "Apply Instagram-style filter",
  edit_type: "style_transfer",
  batch_settings: {
    max_concurrent: 3,
    error_handling: "continue_on_error",
  },
});

Development

Contributing

  1. Fork the repository

  2. Create a feature branch

  3. Make your changes

  4. Add tests for new functionality

  5. Submit a pull request

Testing

# Run tests
npm test

# Run tests in watch mode
npm run test:watch

# Run linting
npm run lint

# Type checking
npm run typecheck

Building

# Build for production
npm run build

# Development mode with hot reload
npm run dev

License

MIT License - see the LICENSE file for details.

Support

This server cannot be deployed

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Maintenance

ActivityInactive
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers

Appeared in Searches