Ali-Flux MCP Server

by echozyr2001

140 downloads Not rated yet

About

A simple MCP server for generating images using Ali Cloud's DashScope API, with tools for task management and local image saving.

Explore

- Provides three tools: generate_image, check_task_status, and download_image
- Generates images via Alibaba Cloud DashScope API using text prompts
- Supports optional image parameters: size, seed, and steps
- Checks task status and downloads generated images to local storage
- Configurable save directory and model name via environment variables

- Node.js and npm
- Alibaba Cloud DashScope API key

- DASHSCOPE_API_KEY: Your Alibaba Cloud DashScope API key
- SAVE_DIR: Directory to save generated images (default: ~/Desktop/flux-images)
- MODEL_NAME: DashScope model name (default: flux-merged)
- WORK_DIR: Work directory (default: process.cwd())

generate_image

Generate images using Alibaba Cloud DashScope API

check_task_status

Check image generation task status

download_image

Download generated images and save them locally

- generate_image - Generate images using Alibaba Cloud DashScope API
- Takes prompt as required parameter
- Optional parameters: size, seed, steps
- Submits image generation task to DashScope API

- check_task_status - Check image generation task status
- Takes task_id as required parameter
- Returns the current status of the image generation task

- download_image - Download generated images and save them locally
- Takes task_id as required parameter
- Optional parameter: save_path for custom save location (must be an absolute path)
- Optional parameter: base_dir for resolving relative paths (defaults to WORK_DIR environment variable)
- Downloads all generated images and saves them to the specified directory

中文文档

A Model Context Protocol server for Alibaba Cloud DashScope API

This is a TypeScript-based MCP server that provides functionality to interact with Alibaba Cloud DashScope API for generating images and saving them locally. It demonstrates core MCP concepts by providing:

- Tools for generating images using Alibaba Cloud DashScope API
- Tools for checking task status
- Tools for downloading generated images and saving them locally

Features

Tools

- generate_image - Generate images using Alibaba Cloud DashScope API - Takes prompt as required parameter - Optional parameters: size, seed, steps - Submits image generation task to DashScope API

- check_task_status - Check image generation task status
- Takes task_id as required parameter
- Returns the current status of the image generation task

- download_image - Download generated images and save them locally
- Takes task_id as required parameter
- Optional parameter: save_path for custom save location (must be an absolute path)
- Optional parameter: base_dir for resolving relative paths (defaults to WORK_DIR environment variable)
- Downloads all generated images and saves them to the specified directory

Development

Prerequisites

- Node.js and npm - Alibaba Cloud DashScope API key

Environment Variables

- DASHSCOPE_API_KEY: Your Alibaba Cloud DashScope API key - SAVE_DIR: Directory to save generated images (default: ~/Desktop/flux-images) - MODEL_NAME: DashScope model name (default: flux-merged) - WORK_DIR: Work directory (default: process.cwd())

Setup

Install dependencies:
npm install

Build the server:

npm run build

For development with auto-rebuild:

npm run watch

Installation

Configuration

To use with Claude Desktop or other MCP-compatible clients, add the server config:

On MacOS: ~/Library/Application Support/Claude/claude_desktop_config.json
On Windows: %APPDATA%/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "ali-flux": {
      "command": "/path/to/ali-flux/build/index.js",
      "env": {
        "DASHSCOPE_API_KEY": "your-api-key-here",
        "SAVE_DIR": "/custom/save/path" // Optional
      }
    }
  }
}

Debugging

Since MCP servers communicate over stdio, debugging can be challenging. We recommend using the MCP Inspector, which is available as a package script:

npm run inspector

The Inspector will provide a URL to access debugging tools in your browser.

No reviews yet — be the first

Sign in to leave a review

Use Google, GitHub, or an email account so ratings stay tied to real people.

Email sign in

No reviews posted yet.