Docs MCP Integration

MCP Integration

Deploy and manage Functions directly from AI assistants using the Model Context Protocol. No context switching required.

Works with:

Claude

Cursor

Windsurf

VS Code

View on GitHub Get API Key

What is MCP?

The Model Context Protocol (MCP) is an open standard that enables AI assistants to interact with external tools and services.

With RocketStack's MCP server, you can deploy, manage, and invoke Functions directly from Claude, Cursor, or any MCP-enabled client — just describe what you want, and the AI handles the rest.

Example: Ask Claude to "deploy my webhook handler" and it will bundle your code and deploy it to RocketStack automatically.

Quick Setup

Get your API key

Create a free account and generate an API key from the console.

Get API Key

Add to your MCP config

Add this to your Claude, Cursor, or other MCP client configuration:

mcp_config.json

{
  "mcpServers": {
    "rocketstack": {
      "command": "npx",
      "args": ["-y", "@rocketstack/mcp-server"],
      "env": {
        "ROCKETSTACK_API_KEY": "rs_live_xxxx"
      }
    }
  }
}

Start deploying

Restart your AI client and start asking it to deploy functions. That's it!

Example Prompts

Natural language prompts your AI will understand:

"Deploy my webhook handler to RocketStack"

Creates and deploys a function

"Show me all my deployed functions"

Lists functions with status

"Test my payment-processor with this payload"

Invokes and returns result

"Show recent logs for email-sender"

Fetches function logs

Available Tools

functions.deploy

Deploy a new function or update existing

functions.list

List all deployed functions

functions.get

Get details of a specific function

functions.invoke

Invoke a function with a payload

functions.delete

Delete a deployed function

functions.logs

Fetch recent logs for a function

Next Steps

Getting Started

Set up your account

Functions Docs

Learn about Functions API