# integromat/make-mcp-server [Health: Active]

**Category:** 🛠️ Other Tools and Integrations  
**Repository:** https://github.com/integromat/make-mcp-server  
**GitHub Stars:** 172  
**npm Downloads (last month):** 390  
**Views:** 3  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/integromat-make-mcp-server

## Description
Turn your Make scenarios into callable tools for AI assistants.

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "make-mcp-server": {
    "command": "npx",
    "args": ["-y","@makehq/mcp-server"],
    "env": {
      "MAKE_API_KEY": "",
      "MAKE_ZONE": "",
      "MAKE_TEAM": ""
    }
  }
}
```

**Requires environment variables:** `MAKE_API_KEY`, `MAKE_ZONE`, `MAKE_TEAM` — the values above are empty placeholders; fill in real credentials before running (see the repository for what each one is for).

## Documentation

## What integromat/make-mcp-server MCP server does

The integromat/make-mcp-server MCP server exposes selected Make automation workflows as tools that an AI assistant can call. It works with scenarios using Make’s On-Demand scheduling, rather than publishing every scenario in an account automatically.

The server connects to a Make account, finds the eligible scenarios, and makes their callable operations available through MCP. This lets an assistant use existing Make logic without requiring that logic to be rebuilt as native agent code. Make remains responsible for the workflow and its connected automation steps, while the MCP client handles tool discovery and invocation.

The repository describes this implementation as legacy. Its README recommends Make’s newer cloud-based MCP server for most use cases, so this project is primarily relevant when a local or existing deployment specifically requires the legacy package.

## How it works

At startup, the server uses Make account credentials and location details to access the relevant team. It identifies scenarios with On-Demand scheduling and examines their input definitions. Those inputs are resolved into parameter descriptions that an AI assistant can use when deciding how to call a scenario.

When the assistant invokes a scenario, the server passes the supplied parameters to Make. The result is returned as structured JSON, allowing the client to process scenario output instead of receiving only unstructured text. This supports workflows where an assistant both sends information into Make and reads the resulting scenario output.

The integromat/make-mcp-server MCP server is intended for MCP clients such as Claude Desktop. The provided configuration runs it locally through Node.js and npx rather than connecting to a hosted endpoint.

## Setup and configuration

The documented setup requires Node.js, an MCP client, and a Make API key with both `scenarios:read` and `scenarios:run` scopes. Add the server to the client’s MCP configuration and run the `@makehq/mcp-server` package with npx.

Three environment variables are needed:

- `MAKE_API_KEY`: an API key generated from the Make profile.
- `MAKE_ZONE`: the hosting zone for the organization, such as an EU Make zone.
- `MAKE_TEAM`: the team identifier, available in the URL of the Make Team page.

The API key must be able to read scenario definitions and run scenarios. The zone and team values determine which Make organization context the server uses.

## Tools and capabilities

Available tools are based on the account’s On-Demand scenarios rather than on a fixed, predefined tool list. The server can:

- Discover eligible Make scenarios.
- Expose scenario inputs with descriptive parameters.
- Trigger scenarios with assistant-provided values.
- Return scenario results as structured JSON.

This model is suitable for exposing existing automation workflows to assistants while keeping the workflow implementation in Make.

## Limitations and notes

Only scenarios configured for On-Demand scheduling are described as discoverable by the server. Access also depends on the Make API key having the required scopes and on correct zone and team settings.

The repository is labeled legacy, and its README directs most users toward Make’s newer cloud-based MCP server. Evaluate that newer option before adopting this package for a new deployment. The supplied material does not specify Make plan requirements, scenario execution limits, or support for MCP clients beyond the Claude Desktop example.

_Full upstream README: https://allmcps.com/mcp/integromat-make-mcp-server/readme_