piapi-mcp-server

What is piapi-mcp-server?

piapi-mcp-server is a TypeScript implementation of a Model Context Protocol (MCP) server that integrates with PiAPI’s API. It allows users to generate media content (images, video, music, 3D models) directly from Claude or any MCP-compatible app, using services like Midjourney, Flux, Kling, LumaLabs, Udio, Chrip, and Trellis.

How to use piapi-mcp-server?

Clone the repository, install dependencies with npm install, set your PiAPI API key in a .env file (PIAPI_API_KEY=your_key_here), then build with npm run build. Start the server with npm start. To use with Claude Desktop, add the server configuration to your claude_desktop_config.json file specifying the path to dist/index.js and your API key.

Key features of piapi-mcp-server

• Generate Flux images from text descriptions
• Works with Claude Desktop and other MCP-compatible apps
• Easy setup with Node.js and npm
• Uses PiAPI’s unified API key authentication
• Built in TypeScript with a clean project structure

Use cases of piapi-mcp-server

• Generate AI images directly from a conversation in Claude Desktop
• Incorporate Flux text‑to‑image generation into custom MCP workflows
• Rapidly prototype media generation without switching between tools

FAQ from piapi-mcp-server

What does piapi-mcp-server do that alternatives don’t?

It provides a direct MCP integration for PiAPI’s media generation APIs, enabling image, video, music, and 3D model generation from any MCP‑compatible client like Claude Desktop. Currently only Flux text‑to‑image is implemented; other features are planned.

What are the runtime requirements?

Node.js 16.x or higher and npm or yarn are required. A PiAPI API key (free from piapi.ai) must be provided via environment variable.

Where does the media data live?

The server acts as a proxy to PiAPI’s cloud APIs. Generated content is delivered by the PiAPI service; no media is stored locally by the server.

What transport or authentication does it use?

The server uses standard MCP transport (stdio). Authentication is handled via a PiAPI API key set in the PIAPI_API_KEY environment variable or in the Claude Desktop config’s env block.

Are there any known limitations?

Only Flux image generation from text descriptions is currently available. Other planned features (Midjourney, Kling, Luma, Suno/Udio, Trellis, etc.) are not yet implemented.