Provides full access to the Make.com API, enabling creation, modification, deletion, activation, and execution of scenarios, as well as management of connections, webhooks, data stores (CRUD), and retrieval of teams/organizations.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@MCP Make.com Server create a scenario that monitors a website and sends an alert if it changes"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
A Model Context Protocol (MCP) server that provides full access to the Make.com API, including the ability to create, modify, and delete scenarios - not just run them.
Feature | Support |
List/Get Scenarios | Yes |
Create Scenarios | Yes |
Update Scenarios (Blueprint) | Yes |
Delete Scenarios | Yes |
Activate/Deactivate | Yes |
Run Scenarios | Yes |
Clone Scenarios | Yes |
Get Blueprints | Yes |
Connections Management | Yes |
Webhooks Management | Yes |
Data Stores CRUD | Yes |
Teams/Organizations | Yes |
Related MCP server: Etsy MCP Server
git clone <repo>
cd mcp-make-server
npm install
npm run buildVariable | Required | Description |
| Yes | Your Make.com API token |
| No | API zone: |
Go to Make.com > Profile > API
Create a new token with these scopes:
scenarios:read
scenarios:write
scenarios:run
connections:read
hooks:read
datastores:read
datastores:write
teams:read
organizations:read
Add to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"make": {
"command": "node",
"args": ["/path/to/mcp-make-server/dist/index.js"],
"env": {
"MAKE_API_TOKEN": "your-api-token-here",
"MAKE_ZONE": "eu1"
}
}
}
}Add to .cursor/mcp.json:
{
"mcpServers": {
"make": {
"command": "node",
"args": ["/path/to/mcp-make-server/dist/index.js"],
"env": {
"MAKE_API_TOKEN": "your-api-token-here",
"MAKE_ZONE": "eu1"
}
}
}
}Tool | Description |
| List all scenarios for a team/org |
| Get scenario details |
| Create a new scenario with blueprint |
| Update scenario name/blueprint/scheduling |
| Delete a scenario |
| Turn on a scenario |
| Turn off a scenario |
| Execute a scenario |
| Duplicate a scenario |
| Get full flow definition |
| Get execution history |
Tool | Description |
| List API connections |
| List webhooks |
Tool | Description |
| List data stores |
| Get data store details |
| Create new data store |
| List records |
| Add a record |
Tool | Description |
| List teams in org |
| List all orgs |
| Get authenticated user |
// Blueprint for a simple HTTP -> JSON scenario
const blueprint = JSON.stringify({
name: "My New Scenario",
flow: [
{
id: 1,
module: "http:ActionGetFile",
version: 3,
mapper: {
url: "https://api.example.com/data",
method: "get"
}
},
{
id: 2,
module: "json:ParseJSON",
version: 1,
mapper: {
json: "{{1.data}}"
}
}
]
});
// Create with on-demand scheduling
create_scenario({
teamId: 123,
blueprint: blueprint,
scheduling: '{"type":"on-demand"}'
});Zone | Region |
| EU (Default) |
| EU |
| US |
| US |
# Run in development mode
npm run dev
# Build for production
npm run build
# Run production build
npm startMIT
Activate a scenario so it can run
Inputs: scenarioId (required).
Clone/duplicate a scenario
Inputs: name (required), teamId (required), scenarioId (required), organizationId (required).
Create a new data store
Inputs: name (required), teamId (required), maxSize.
Create a record in a data store
Inputs: data (required), dataStoreId (required).
Create a new scenario with a blueprint. The blueprint defines the workflow modules and their connections.
Inputs: teamId (required), blueprint (required), scheduling (required), name, +1 more.
Deactivate a scenario to stop it from running
Inputs: scenarioId (required).
Delete a scenario permanently
Inputs: scenarioId (required).
Get information about the authenticated user
Inputs: no named parameters.
Get details of a data store
Inputs: dataStoreId (required).
Get details of a specific scenario
Inputs: scenarioId (required).
Get the full blueprint (flow definition) of a scenario
Inputs: scenarioId (required).
Get execution logs for a scenario
Inputs: scenarioId (required), limit.
List all connections (API credentials) for a team
Inputs: teamId (required), limit.
List records in a data store
Inputs: dataStoreId (required), limit.
List all data stores for a team
Inputs: teamId (required), limit.
List all webhooks for a team
Inputs: teamId (required), limit.
List all organizations the user has access to
Inputs: limit.
List all scenarios for a team or organization
Inputs: limit, offset, teamId, organizationId.
List all teams in an organization
Inputs: organizationId (required), limit.
Run a scenario immediately. Scenario must be active and set to on-demand scheduling.
Inputs: scenarioId (required), data, responsive.
Update an existing scenario (name, blueprint, scheduling, folder)
Inputs: scenarioId (required), name, folderId, blueprint, +1 more.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 21 tools
Each tool targets a distinct resource and action (scenario lifecycle, data stores, teams, orgs, user, connections, hooks). The scenario tools are clearly differentiated by verb: get, list, create, update, delete, activate, deactivate, run, clone, get_blueprint, get_logs. No two tools appear to do the same thing.
All tool names use snake_case with a consistent verb_noun pattern (e.g., list_scenarios, create_scenario, get_data_store). There are no mixed conventions or vague verbs. The pattern extends to compound resources like create_data_store_record and list_data_store_records.
21 tools is on the heavy side for the server's scope, falling into the 16-25 borderline range. While most tools earn their place by covering distinct operations, the set could likely be consolidated without losing functionality. The count feels somewhat inflated for the core scenario and data store management tasks.
Scenario management is comprehensive (full CRUD plus activate, deactivate, run, clone, blueprint, logs). However, notable gaps exist: data stores and records lack update/delete operations, and connections and hooks are only listable with no create, update, or delete. These missing lifecycle operations will limit agents needing full management.
Build and run AI workflows, agents, data tables and 1000+ API integrations from any MCP client.
Read MagicBell broadcasts, users, events, workflows and channels; create and update user records.
Read spaces, projects, tasks and comments; create tasks and comments, update tasks in Nozbe Teams.
Read Fillout forms and submissions, export responses, and create submissions and webhooks.
