Skip to main content
Glama
ImageAPI AI MCP Server

ImageAPI AI MCP Server

Official
by ImageAPIAI

ImageAPI AI MCP Server (@imageapiai/mcp)

Official Model Context Protocol (MCP) server for ImageAPI AI. Connects AI coding assistants and agents (such as Cursor, Claude Desktop, and Windsurf) to generate high-resolution images, refine prompts for 0 credits, and inspect account credits directly within your development workflows and chat sessions.

⚡ Quick Start

1. Claude Desktop Setup

Add the following to your claude_desktop_config.json:

  • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json

  • Windows: %APPDATA%\Claude\claude_desktop_config.json


{
  "mcpServers": {
    "imageapiai": {
      "command": "npx",
      "args": ["-y", "@imageapiai/mcp"],
      "env": {
        "IMAGEAPIAI_API_KEY": "sk_live_YOUR_API_KEY_HERE"
      }
    }
  }
}

2. Cursor IDE Setup

Add to your project root under .cursor/mcp.json (or globally in ~/.cursor/mcp.json):


{
  "mcpServers": {
    "imageapiai": {
      "command": "npx",
      "args": ["-y", "@imageapiai/mcp"],
      "env": {
        "IMAGEAPIAI_API_KEY": "sk_live_YOUR_API_KEY_HERE"
      }
    }
  }
}

3. Windsurf Setup

Add to ~/.codeium/windsurf/mcp_config.json:


{
  "mcpServers": {
    "imageapiai": {
      "command": "npx",
      "args": ["-y", "@imageapiai/mcp"],
      "env": {
        "IMAGEAPIAI_API_KEY": "sk_live_YOUR_API_KEY_HERE"
      }
    }
  }
}

Related MCP server: imagegen-mcp

🛠️ Available MCP Tools

Once installed, your AI agent has native access to the following 4 tools:

1. generate_image

Generates a new AI image from scratch based on a primary text prompt. Deducts standard credits from your account balance.

  • Parameters:

  • prompt (string, required): Primary text prompt describing the image to generate.

  • width (number, optional, default: 512): Width in pixels (min: 256, max: 2048).

  • height (number, optional, default: 512): Height in pixels (min: 256, max: 2048).

  • quality (string, optional, default: "medium"): Inference quality preset ("low", "medium", "high").

  • Example Prompt to AI:"Generate a 1024x1024 high quality cyberpunk city street in the rain."

  • Output: Returns JSON containing image_url, prompt_id, credits_remaining, and credits_deducted.

2. refine_image

Refines an existing image by appending modifications to the original prompt. Allows up to 5 free retries per generation ID for 0 credits.

  • Parameters:

  • parent_prompt_id (string, required): The prompt_id or generation_id of the original base image.

  • prompt_update (string, required): Additions or modifications to append to the base prompt.

  • quality (string, optional, default: "medium"): Optional quality level override ("low", "medium", "high").

  • Example Prompt to AI:"Take image gen_a1b2c3d4e5f6 and refine it: add golden neon reflections on the ground."

  • Output: Returns JSON containing the updated image_url, retries_remaining, and effective_prompt.

3. get_profile

Fetches user account details, current subscription status, and remaining monthly and purchased credit balance.

  • Parameters: None

  • Example Prompt to AI:"Check how many ImageAPI credits I have left."

  • Output: Returns JSON with email, credit_balance, monthly_credits, purchased_credits, and subscription_status.

4. get_history

Retrieves a list of recent historical image generations and their CDN asset URLs associated with the authenticated account.

  • Parameters: None

  • Example Prompt to AI:"List my latest 5 generated images from ImageAPI."

  • Output: Returns an array of historical image objects containing generation_id, original_prompt, prompt_text, r2_image_url, and retry_number.

🔑 Environment Variables

Variable

Description

IMAGEAPIAI_API_KEY

Your secret API key (sk_live_...). Obtainable from imageapiai.com/dashboard.

IMAGEAPI_API_KEY

Supported alias fallback.

NEXT_PUBLIC_IMAGEAPIAI_API_KEY

Supported frontend/Next.js environment fallback.

🔗 Resources


📄 License

MIT © ImageAPI AI

Available Tools

4 tools
generate_imageA

Generate a fresh high-resolution AI image from a text prompt. Deducts credits from your ImageAPI account.

Inputs: prompt (required), width, height, quality.

get_historyA

Retrieve a list of all historical AI image generations associated with your ImageAPI account.

Inputs: no named parameters.

get_profileA

Fetch your ImageAPI account profile, credit balance, monthly allowance, and active subscription status.

Inputs: no named parameters.

refine_imageA

Refine or modify an existing generated image for 0 credits (up to 5 free retries per parent prompt ID). Appends prompt modifications to the locked base prompt.

Inputs: prompt_update (required), parent_prompt_id (required), width, height, +1 more.

Tool Schema Changelog

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

Loading tool schema changelog…

    TDQS

    A4.2/5.0

    Scored across 4 tools

    Disambiguation5/5

    Each tool has a clearly distinct purpose: generate creates new images, refine modifies existing ones, get_profile retrieves account data, and get_history lists past generations. No overlap or ambiguity between tools.

    Naming Consistency5/5

    All tool names follow a consistent verb_noun pattern: generate_image, refine_image, get_profile, get_history. This makes the API predictable and easy to navigate.

    Tool Count5/5

    With only 4 tools, the server is well-scoped for its purpose. Each tool covers a core capability without redundancy or bloat.

    Completeness5/5

    The surface covers the full lifecycle of image generation (create, refine) plus account management (profile, history). No obvious gaps for an image generation API.

    Maintenance

    ActivityInactive
    ResponsivenessOutdated data

    Related MCP Connectors

    Related MCP Servers