satori
An MCP server that uses Satori to generate consistent images from React components.
claude mcp add --transport stdio jellypod-inc-satori-mcp-server node server.js \ --env BLOB_READ_WRITE_TOKEN="your-blob-storage-token"
How to use
Satori MCP Server lets you generate beautiful images from React components using the Satori renderer. It exposes MCP endpoints that accept JSX-like content or pre-built templates and returns references to generated images stored in Vercel Blob Storage. The built-in tools include generate_image for creating images from custom JSX, and generate_from_template for applying predefined templates (social cards, blog headers, and quotes) with parameters such as title, description, and background color. You can also list available templates with list_templates. This makes it easy to produce social-ready images for blog posts, quotes, or hero headers directly from an AI assistant or automation workflow.
How to install
Prerequisites:
- Node.js 20+
- pnpm (recommended) or npm
Installation steps:
- Clone the repository git clone https://github.com/Jellypod-Inc/satori-mcp-server.git
- Navigate into the project cd satori-mcp-server
- Install dependencies
pnpm install
or npm install
- Run the development server
pnpm dev
or npm run dev
Notes:
- The server stores generated images to Vercel Blob Storage; ensure you provide a valid BLOB_READ_WRITE_TOKEN in the environment.
- For production, build and then start the production server: pnpm build pnpm start
Additional notes
Environment variables and configuration tips:
- BLOB_READ_WRITE_TOKEN: Required for saving and accessing generated images on Vercel Blob Storage. Obtain from the Vercel dashboard and set it in your environment.
- The server is intended to be run behind MCP clients (e.g., Claude Desktop) or via MCP tooling; ensure your MCP client is configured to point to the server URL when using hosted mode.
- Templates support googleFonts for font loading; you can specify fonts in the template params or image generation calls to ensure consistent typography. Common issues:
- If images fail to generate, verify that the blob storage token is valid and that the network can reach the Vercel Blob Storage endpoint.
- Ensure Node.js version matches the project requirements (Node 20+).
- For large images, MCP responses may include a storage URL rather than a base64 image; handle the URL returned by the MCP client accordingly.
Related MCP Servers
iterm
A Model Context Protocol server that executes commands in the current iTerm session - useful for REPL and CLI assistance
mcp
Octopus Deploy Official MCP Server
furi
CLI & API for MCP management
editor
MCP Server for Phaser Editor
DoorDash
MCP server from JordanDalton/DoorDash-MCP-Server
mcp
MCP сервер для автоматического создания и развертывания приложений в Timeweb Cloud