Skip to main content
Glama

Ideogram MCP Server

License: MIT Node.js Version npm version

A Model Context Protocol (MCP) server implementation for the Ideogram AI API. This server provides tools for generating, editing, and describing images using Ideogram's powerful AI models.

Features

  • Generate images from text descriptions

  • Edit existing images using masks

  • Generate descriptions for images

  • Download generated images to local directory

  • Prompt templates for common operations

Related MCP server: Replicate Ideogram V3 MCP Server

Installation

  1. Clone the repository

git clone https://github.com/flowluap/ideogram-mcp-server.git
cd ideogram-mcp-server
  1. Install dependencies

npm install
  1. Build the project

npm run build

Cursor Integration

To use this MCP in Cursor, add the following configuration to your Cursor settings (Settings > AI > Model Context Protocol):

{
  "mcps": {
    "ideogram-mcp": {
      "command": "node",
      "args": ["PATH_TO_YOUR_DIST/index.js"],
      "env": {
        "IDEOGRAM_API_KEY": "your-api-key-here"
      }
    }
  }
}

Replace:

  • PATH_TO_YOUR_DIST/index.js with the absolute path to your built index.js file

  • your-api-key-here with your actual Ideogram API key

Example (redacted):

{
  "mcps": {
    "ideogram-mcp": {
      "command": "node",
      "args": ["/path/to/ideogram-mcp-server/dist/index.js"],
      "env": {
        "IDEOGRAM_API_KEY": "xxxxx-REDACTED-xxxxx"
      }
    }
  }
}

Available Tools

  1. generate

    • Generate images from text prompts

    • Parameters:

      • prompt: string (required)

      • aspect_ratio: string (optional) - One of: ASPECT_1_1, ASPECT_16_9, ASPECT_9_16, ASPECT_4_3, ASPECT_3_4

      • model: string (optional) - One of: V_1, V_2, V_2_TURBO

      • magic_prompt_option: string (optional) - One of: AUTO, ON, OFF

      • num_images: number (optional) - Between 1 and 8

      • seed: number (optional) - Between 0 and 2147483647

  2. edit

    • Edit images using masks

    • Parameters:

      • image_file: string (required) - Base64 or file path

      • mask: string (required) - Base64 or file path

      • prompt: string (required)

      • model: string (required) - One of: V_2, V_2_TURBO

  3. describe

    • Generate descriptions for images

    • Parameters:

      • image_file: string (required) - Base64 or file path

  4. download_images

    • Download generated images to local directory

    • Parameters:

      • urls: string[] (required) - Array of image URLs

      • output_dir: string (required) - Output directory path

Prompt Templates

  1. generate_image

    • Template for generating images

    • Parameters:

      • description: string

Development

# Build the project
npm run build

# Run in development mode
npm run dev

# Run tests
npm test

# Run linter
npm run lint

Requirements

  • Node.js >= 18.0.0

  • Ideogram API key

License

This project is licensed under the MIT License - see the LICENSE file for details.

Contributing

  1. Fork the repository

  2. Create your feature branch

  3. Commit your changes

  4. Push to the branch

  5. Create a new Pull Request

Available Tools

4 tools
describeD

Inputs: image_file (required).

download_imagesD

Inputs: urls (required), output_dir (required).

editD

Inputs: image_file (required), mask (required), prompt (required), model (required).

generateD

Inputs: prompt (required), aspect_ratio, model, magic_prompt_option, +2 more.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    D1.5/5.0

    Scored across 4 tools

    Disambiguation2/5

    Tool names are distinct (describe, download_images, edit, generate) but without descriptions, an agent cannot reliably determine their boundaries. For example, 'describe' might overlap with 'generate' if generating a description, leading to potential misselection.

    Naming Consistency2/5

    Naming is inconsistent: 'describe', 'edit', and 'generate' are single verbs, while 'download_images' uses a verb_noun pattern with an underscore. No consistent verb_noun or camelCase pattern is followed.

    Tool Count3/5

    4 tools is within the typical well-scoped range (3-15), but for an image generation server, the set feels slightly thin. Still, each tool appears to serve a distinct purpose without being excessive.

    Completeness3/5

    The tools cover basic operations: generate, edit, describe, and download images. However, obvious gaps exist, such as missing list, delete, or style-transfer tools, which may cause agent failures in common workflows.

    Maintenance

    ActivityInactive
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers