Ali-Flux MCP Server

What is Ali-Flux MCP Server?

It is a TypeScript-based Model Context Protocol server that integrates with Alibaba Cloud DashScope API to generate images and save them locally. It is intended for developers using MCP-compatible clients (e.g., Claude Desktop) who need AI image generation capabilities.

How to use Ali-Flux MCP Server?

Install dependencies with npm install, build the server with npm run build, and set the required environment variables (DASHSCOPE_API_KEY, optionally SAVE_DIR, MODEL_NAME, WORK_DIR). Then configure your MCP client (e.g., Claude Desktop) using the command and env fields as shown in the README.

Key features of Ali-Flux MCP Server

• 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

Use cases of Ali-Flux MCP Server

• Generate images from text prompts directly within Claude Desktop
• Automate image creation pipelines with task status polling
• Save generated images to a custom local folder for later use
• Integrate Alibaba Cloud image generation into personal or business workflows

FAQ from Ali-Flux MCP Server

What tools does Ali-Flux MCP Server provide?

It offers generate_image (submit a prompt and optional parameters), check_task_status (poll for task completion), and download_image (save generated images to a local directory).

What environment variables are required?

You must set DASHSCOPE_API_KEY. Optional variables include SAVE_DIR (default: ~/Desktop/flux-images), MODEL_NAME (default: flux-merged), and WORK_DIR (default: current working directory).

How do I debug the server?

Since MCP servers communicate over stdio, use the built-in script npm run inspector which launches the MCP Inspector for debugging.

What image parameters can I customize?

When calling generate_image, you can specify size, seed, and steps in addition to the required prompt.

How are generated images saved?

The download_image tool saves all images from a completed task to the directory specified by SAVE_DIR or a custom save_path (must be absolute). A base_dir can be used for resolving relative paths.