# Make (Integromat)

> Trigger Make scenarios and inspect runs from an AI agent.

[Canonical HTML page](https://top-mcps.com/mcp/make) · [server.json](https://top-mcps.com/mcp/make.json) · [methodology](https://top-mcps.com/about/methodology)

## Install

### Claude Desktop — `claude_desktop_config.json`

Paste under mcpServers. Fully quit and reopen Claude after editing.

```json
{
  "mcpServers": {
    "make": {
      "command": "npx",
      "args": [
        "-y",
        "make-mcp-server"
      ],
      "env": {
        "MAKE_API_TOKEN": "${MAKE_API_TOKEN}",
        "MAKE_ZONE": "${MAKE_ZONE}"
      }
    }
  }
}
```

### Claude Code — `CLI or .mcp.json`

Run from your repo. Commit .mcp.json to share with your team.

```shell
# export MAKE_API_TOKEN=your-make-api-token
# export MAKE_ZONE=eu1.make.com
claude mcp add make -- npx -y make-mcp-server
```

### Cursor — `.cursor/mcp.json`

Global path: ~/.cursor/mcp.json. Reload window after editing.

```json
{
  "mcpServers": {
    "make": {
      "command": "npx",
      "args": [
        "-y",
        "make-mcp-server"
      ],
      "env": {
        "MAKE_API_TOKEN": "${MAKE_API_TOKEN}",
        "MAKE_ZONE": "${MAKE_ZONE}"
      }
    }
  }
}
```

### VS Code — `.vscode/mcp.json`

VS Code uses the "servers" key (not "mcpServers").

```jsonc
{
  "servers": {
    "make": {
      "command": "npx",
      "args": [
        "-y",
        "make-mcp-server"
      ],
      "env": {
        "MAKE_API_TOKEN": "${MAKE_API_TOKEN}",
        "MAKE_ZONE": "${MAKE_ZONE}"
      }
    }
  }
}
```

### Windsurf — `~/.codeium/windsurf/mcp_config.json`

Open via Cascade → hammer icon → Configure.

```json
{
  "mcpServers": {
    "make": {
      "command": "npx",
      "args": [
        "-y",
        "make-mcp-server"
      ],
      "env": {
        "MAKE_API_TOKEN": "${MAKE_API_TOKEN}",
        "MAKE_ZONE": "${MAKE_ZONE}"
      }
    }
  }
}
```

### Cline — `cline_mcp_settings.json`

Open via the Cline sidebar → MCP Servers → Edit.

```json
{
  "mcpServers": {
    "make": {
      "command": "npx",
      "args": [
        "-y",
        "make-mcp-server"
      ],
      "env": {
        "MAKE_API_TOKEN": "${MAKE_API_TOKEN}",
        "MAKE_ZONE": "${MAKE_ZONE}"
      }
    }
  }
}
```

### Continue — `~/.continue/config.json`

Continue uses modelContextProtocolServers with a transport block.

```json
{
  "experimental": {
    "modelContextProtocolServers": [
      {
        "transport": {
          "type": "stdio",
          "command": "npx",
          "args": [
            "-y",
            "make-mcp-server"
          ],
          "env": {
            "MAKE_API_TOKEN": "${MAKE_API_TOKEN}",
            "MAKE_ZONE": "${MAKE_ZONE}"
          }
        }
      }
    ]
  }
}
```

### Codex CLI — `~/.codex/config.toml`

Codex uses TOML. Each server is a [mcp_servers.<name>] subtable.

```shell
# ~/.codex/config.toml
[mcp_servers.make]
command = "npx"
args = [
  "-y",
  "make-mcp-server",
]
env = { MAKE_API_TOKEN = "${MAKE_API_TOKEN}", MAKE_ZONE = "${MAKE_ZONE}" }
```

### Zed — `~/.config/zed/settings.json`

Zed calls them "context_servers". Settings live-reload on save.

```jsonc
{
  "context_servers": {
    "make": {
      "command": {
        "path": "npx",
        "args": [
          "-y",
          "make-mcp-server"
        ]
      },
      "env": {
        "MAKE_API_TOKEN": "${MAKE_API_TOKEN}",
        "MAKE_ZONE": "${MAKE_ZONE}"
      }
    }
  }
}
```

### ChatGPT — `ChatGPT → Apps directory`

Make (Integromat) doesn't ship a hosted HTTPS endpoint today. ChatGPT supports remote MCP servers only — to use this server in ChatGPT you'll need to deploy it to a public HTTPS URL first (e.g. via Cloudflare Workers or Vercel) or wait for an official remote build.

```none

```

## At a glance

- **Maintainer:** Community (unofficial Make.com MCP)
- **Transport:** stdio
- **Auth model:** API key
- **Required secrets:** MAKE_API_TOKEN, MAKE_ZONE
- **Supported clients:** Claude, Cursor, VS Code, Windsurf, Any MCP-compatible client, Make Cloud (any region)
- **License:** MIT
- **Language:** TypeScript
- **Latest version:** latest
- **Last verified:** 2026-05-31
- **GitHub stars:** 158 (fetched 2026-06-02T11:55:52.354Z)
- **Score:** 70/100 (rubric 2026-04 — see https://top-mcps.com/about/methodology)
- **Source:** https://github.com/integromat/make-mcp-server

## Security & scope

- **Access scope:** network
- **Sandbox:** Make API token in env. Tokens are scoped to one organisation. Create a dedicated team member account for the MCP with access only to the scenarios it needs to trigger.
- **Gotchas:**
  - Make tokens have broad scope by default. Create a non-admin team-member account for the MCP.
  - Triggering paid-tier scenarios costs operations on every run — be aware of per-trigger cost.
  - Audit log per scenario is the forensic trail; review it after any agent-triggered run.

## Quick answer

**What it does.** Connects to a Make organisation, lists scenarios, runs them on demand, and reads execution status and run logs.

**Best for:**
- Triggering existing Make scenarios
- Inspecting failed runs
- Listing scenarios available to the agent
- Integrating Make with other agent workflows

**Not for:**
- Authoring scenarios (use the Make visual builder)
- High-frequency triggers (use Make webhooks directly)

## Description

A community-maintained MCP that wraps the Make (formerly Integromat) API so an AI agent can list scenarios, run them on demand, and inspect execution status. Make is the most visually-oriented of the major automation platforms; the MCP gives agents the same trigger surface that the Make UI exposes.

## Why it matters

Make sits in a lot of mid-market stacks as the "Zapier for power users." An MCP wrapper turns the visual scenarios into agent-callable tools.

## Key features

- Make Cloud + EU + US regions
- API-token auth scoped to one org
- Scenario enumeration
- On-demand execution
- Run-log inspection

## FAQ

### Which Make regions?

All — pass MAKE_ZONE (eu1.make.com, us1.make.com, etc.) to match your org's region. Cross-region API calls do not work.

### Does the agent see scenario data?

Only run-log output. Internal scenario step data may include sensitive values — review which scenarios you expose before granting access.

### Can the agent author new scenarios?

No. Scenarios are authored in the Make visual editor; the MCP only triggers and inspects.

### Free plan support?

Yes — the API is available on Make Free, but with low rate limits. For agent-driven use on a team, expect to need a paid plan.

## Changelog

- **2026-05-31** — Refreshed install snippets and fact sheet; verified for 2026.
- **2025-05-10** — Initial directory listing.