@varity-labs/mcp

The published Varity MCP package for Cursor, Claude Code, VS Code, Windsurf, and other MCP clients.

This repository contains only @varity-labs/mcp, the npm MCP server. It is a thin tool wrapper around varitykit and Varity's gateway APIs; it is not the Python CLI, the portal, the dormant SDK, or an app-store package.

The Varity MCP Server lets your AI editor build, deploy, and manage supported apps in production for you. Flat monthly cost per app, locked at deploy time based on hardware reserved — no usage-based billing, no surprise overages. One server, every AI client, zero commands.

Browser usage: see the browser usage guide for Claude.ai or ChatGPT browser. Quick start: pick your editor below and run one command.

Install

Cursor

Add to .cursor/mcp.json:

{
  "mcpServers": {
    "varity": {
      "command": "npx",
      "args": ["-y", "@varity-labs/mcp"]
    }
  }
}

Claude Code

claude mcp add varity -- npx -y @varity-labs/mcp

VS Code with Copilot

1. Command Palette → MCP: Add Server
2. Select Command (stdio)
3. Command: npx -y @varity-labs/mcp
4. Name: Varity

Windsurf

Add to mcp_config.json:

{
  "mcpServers": {
    "varity": {
      "command": "npx",
      "args": ["-y", "@varity-labs/mcp"]
    }
  }
}

Claude.ai / ChatGPT (HTTP)

The hosted process is reachable at:

https://mcp.varity.so

Hosted OAuth and per-user deployment authorization are not currently certified end to end. Do not use this endpoint for owner-scoped production operations yet; use the stdio package until the repository architecture's OAuth verification bar passes.

Any MCP-compatible client (stdio)

{
  "mcpServers": {
    "varity": {
      "command": "npx",
      "args": ["-y", "@varity-labs/mcp"]
    }
  }
}

What you can do from your AI editor

The MCP exposes Varity's deploy surface so your AI tool can take action — not just generate code. Try prompts like:

Deploy your own code

• "Deploy this project to production"
• "Ship this app live"
• "What would it cost to deploy this on Varity?"

Deploy a certified template (one command, no code required)

• "What templates can Varity deploy?"
• "Spin up Agent Zero"
• "Deploy a private app template for me"

Manage live deployments

• "Show my deployments"
• "What's the URL of my last deploy?"
• "Stop my-app and stop billing it"
• "Get the build logs for my deployment"

Migrate from Vercel

• "Migrate my Vercel app at github.com/me/my-app to Varity"
• "Preview what changes the migration will make"

Docs and pricing

• "Search Varity docs for environment variables"
• "How much would it cost to host a 5,000-user API on Varity?"
• "What's my monthly cost going to be if my app gets traction?"

Tools

Templates

Varity templates come from the gateway-owned certified catalog. Ask your AI editor "what templates can I deploy?" or "deploy Agent Zero for me" and it will list the live catalog, inspect that template contract, and deploy it through varitykit app deploy --template <id>.

Each template reserves different hardware. Use varity_template_info to see the required environment variables, private/public access mode, resources, hardware profile, and certification state before deploying.

End-to-end example

From empty folder to deployed app, all in natural language:

You: "Make me a simple landing page for my coffee shop and deploy it"
AI:  Wrote the landing page, ran the build, deployed live at
     https://varity.app/coffee-shop/

You: "Now deploy Agent Zero"
AI:  Agent Zero is certified and does not require environment variables.
AI:  Deployed Agent Zero at https://varity.app/my-agent/

How Varity is priced

• Flat monthly cost per app: locked at deploy time, based on the hardware your app reserves. Your bill on day 1 equals your bill on day 1000.
• No usage-based billing: cost doesn't change with traffic, requests, bandwidth, or build minutes. No surprise overages.
• No plan tiers: every feature is in every account; you don't pay for "Pro" to unlock anything.

Ask your AI editor "how much would this app cost on Varity?" and it will use varity_cost_calculator to estimate before you deploy.

Transports

stdio (default)

For desktop AI editors. Cursor, Claude Code, VS Code, Windsurf.

npx -y @varity-labs/mcp

HTTP

For browser-based AI tools. Claude.ai, ChatGPT.

npx -y @varity-labs/mcp --transport http --port 3100

The process is hosted at https://mcp.varity.so, but authenticated hosted use is not currently certified. Use stdio for owner-scoped operations.

Prerequisites

• Node.js >= 18
• For deployment: pip install varitykit

Cost

Varity charges a flat monthly price per app, locked at deploy time based on the hardware reserved. Your bill doesn't grow with your traffic, requests, or build minutes. Use the varity_cost_calculator tool from your AI editor for a detailed estimate before you deploy.

Deploy supported apps from your AI coding tool. Flat predictable pricing — your bill stays the same as your app grows.

Documentation · GitHub · Discord