Get the FREE Ultimate OpenClaw Setup Guide →

MeiGen-AI-Design

MeiGen-AI-Design-MCP — Turn Claude Code / OpenClaw into your local Lovart. Local ComfyUI, 1,400+ prompt library, multi-direction parallel generation.

Installation
Run this command in your terminal to add the MCP server to Claude Code.
Run in terminal:
Command
claude mcp add --transport stdio jau123-meigen-ai-design-mcp npx -y meigen@latest \
  --env MEIGEN_API_TOKEN="meigen_sk_..."

How to use

MeiGen AI Design MCP is a feature-rich design-oriented MCP server that integrates with local ComfyUI workflows and other MCP hosts. It exposes a suite of tools and slash commands to help LLMs craft, organize, and execute complex image generation tasks across multiple providers, with parallel execution and prompt engineering baked in. Core capabilities include searching a large gallery of prompts, retrieving inspiration with detailed metadata, enhancing prompts for production-ready outputs, listing models across configured providers, uploading reference images, and managing ComfyUI workflows. At generation time, the server can automatically route requests to the best available provider, streamlining the creation of multi-directional outputs and parallel image batches. You can trigger quick image generation with a single slash command or engage the full prompt-crafting workflow for batch tasks.

How to install

Prerequisites:

  • Node.js (16.x or newer) and npm installed on your system
  • Access to a host where MCP servers are consumed (e.g., Claude Code environment, or an MCP-compatible host)

Installation steps:

  1. Install the MeiGen MCP server into your MCP ecosystem via the marketplace or npm integration (example for MCP config usage):
# If using Claude Code plugin marketplace (recommended path):
/plugin marketplace add jau123/MeiGen-AI-Design-MCP
/plugin install meigen@meigen-marketplace
  1. Add MeiGen to your MCP configuration (example config):
{
  "mcpServers": {
    "meigen": {
      "command": "npx",
      "args": ["-y", "meigen@latest"],
      "env": {
        "MEIGEN_API_TOKEN": "meigen_sk_..."
      }
    }
  }
}
  1. If you are using a different MCP host, ensure the server entry is included in your .mcp.json or equivalent config file, then refresh/reload your MCP environment.

  2. Start consuming the server through your MCP-enabled tooling. For example, after installation, the host will invoke: npx meigen@latest (as configured) to run the MeiGen MCP server.

Prerequisites recap:

  • Node.js and npm
  • Access to MCP-compatible host (Claude Code, OpenClaw, etc.)
  • Optional: MEIGEN_API_TOKEN for authenticated features

Additional notes

Tips and notes:

  • The MEIGEN_API_TOKEN env var enables API-based features; you can start with free features such as inspiration search and prompt enhancement without a key.
  • MeiGen can operate with local ComfyUI (no external API required) or with cloud/open APIs by configuring the provider in the setup wizard.
  • When using multiple providers, MeiGen will route generate tasks to the best available provider and leverage parallel sub-agents for faster batch generation.
  • Typical issues involve network access to local ComfyUI or misconfigured API tokens. Ensure the variable names in your MCP config match exactly and that tokens are valid.
  • You can customize behavior with the provided slash commands and MCP tools (search_gallery, generate_image, comfyui_workflow, etc.).

Related MCP Servers

Sponsor this space

Reach thousands of developers