Make MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| MAKE_ZONE | Yes | Region (zone) of your Make instance, e.g., eu1, us1, etc. Default is eu1. | eu1 |
| MAKE_ORG_ID | No | Default organization ID (optional) | |
| MAKE_TEAM_ID | No | Default team ID (optional) | |
| MAKE_API_BASE | No | Full override of base URL (optional) | |
| MAKE_API_TOKEN | Yes | API token for Make.com (required) |
Capabilities
Features and capabilities supported by this server
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| make_list_organizationsA | List all Make.com organizations the API token has access to. Returns id + name. Use this first if you don't know the organization_id. |
| make_list_teamsA | List teams inside a Make.com organization (id + name). teamId is required by make_list_scenarios. If organization_id is omitted, the configured one is used, or the single org is auto-selected. |
| make_list_scenariosA | List Make.com scenarios (automations) for a team or organization. Provide team_id OR organization_id (falls back to the configured ones, or the single org). Set active_only=true to return only active scenarios. Results are capped (default 500); 'capped' tells you if more exist. |
| make_get_scenarioA | Get full details of a single Make.com scenario by its ID (status, scheduling, team, description, etc.). |
| make_get_blueprintA | Get the full blueprint (JSON structure) of a Make.com scenario — every module, its parameters, filters and how modules connect to each other. This is the same data Make's own editor renders as the canvas. Use it to inspect exact module configuration without asking the user for screenshots. |
| make_clone_scenarioA | Clone a Make.com scenario into a new one with a given name — the safe way to test structural changes (e.g. before make_update_blueprint) without touching the live scenario. team_id/organization_id fall back to the configured ones or the single team/org. states=true also copies module execution state (e.g. last trigger position); default false gives a clean copy. |
| make_update_blueprintA | HIGH RISK: update a scenario's blueprint (module structure/params/filters/connections), scheduling, name, and/or folder via PATCH — modifies REAL scenario structure. MANDATORY workflow before calling: (1) call make_get_blueprint to see the current structure, (2) present the user the exact planned change/diff and get their explicit approval, (3) strongly prefer testing structural changes on a clone first (make_clone_scenario) before touching the live scenario. Requires confirmed:true as proof this workflow happened (this tool's own safety gate — Make's API does not require it; refuses otherwise). Before applying, takes a fresh snapshot of the current blueprint and returns it alongside the update result as 'before_blueprint', so the prior state can be manually restored (by calling this tool again with that blueprint) if the change turns out wrong — this is a recovery aid only, it does not block or compare against that snapshot. |
| make_run_scenarioA | Run a Make.com scenario on demand (right now). By default waits for the run to finish and returns the result (Make waits up to ~40s); set wait=false to trigger and return the executionId immediately without waiting. 'data' is optional input passed to the scenario (only relevant if it starts with a trigger that accepts input). NOTE: the scenario must be ACTIVE — Make returns HTTP 422 'Scenario is not activated' for an inactive scenario; if needed, activate it first with make_start_scenario. |
| make_start_scenarioA | Activate (turn ON / schedule) a Make.com scenario so it runs on its schedule or trigger. |
| make_stop_scenarioA | Deactivate (turn OFF) a Make.com scenario so it stops running on its schedule/trigger. |
| make_list_executionsA | List the execution history (logs) of a Make.com scenario, newest first. Shows whether runs succeeded/failed and when. Optionally filter by status (1=success, 2=warning, 3=error). |
| make_get_executionA | Get full details of a single scenario execution/log (status, operations, duration, error info) by scenario_id + execution_id (from make_list_executions). |
| make_open_in_browserA | Open a Make.com URL (e.g. a scenario editor link) in the default browser on the user's computer. |
| make_list_data_storesA | List Make.com data stores (Make's built-in lightweight database) for a team. team_id falls back to the configured team or the only team. Returns id, name and record info. |
| make_get_recordsA | List the records in a Make.com data store. Each record has a 'key' and a 'data' object. Results are capped (default 100). |
| make_add_recordA | Add a new record to a Make.com data store. 'data' is an object with the record's fields (must match the data store's structure). 'key' is optional — Make auto-generates one if omitted. |
| make_update_recordA | Replace an existing record in a Make.com data store, identified by its key. 'data' is the full new object for the record (it REPLACES the record). |
| make_delete_recordA | Delete SPECIFIC record(s) from a Make.com data store by key(s). Pass 'keys' as an array of the record keys to delete — ONLY those are removed. Deleting ALL records is intentionally NOT supported here, for safety. |
| make_list_incompleteA | List incomplete/failed executions (DLQ) of a Make.com scenario — runs that errored and are waiting. Use this before retrying. |
| make_retry_executionA | Retry a single incomplete/failed execution (DLQ item) by its id (from make_list_incomplete). This RE-RUNS the failed execution. |
| make_get_dlq_bundleA | Get the actual data (bundle) that was being processed at the moment a scenario execution failed — the real input/output values behind an incomplete/failed execution (from make_list_incomplete). Use this for real diagnosis instead of guessing from error messages alone. |
| make_delete_dlqA | Delete specific incomplete/failed execution(s) (DLQ items) by id, from a given scenario. Pass 'dlq_ids' as an array of the exact ids to delete (from make_list_incomplete) — ONLY those are removed. Deleting ALL incomplete executions of a scenario is intentionally NOT supported here, for safety. |
| make_list_hooksA | List Make.com webhooks for a team. Returns each hook's id, name and its webhook URL (used by make_trigger_webhook). team_id falls back to the configured team or the only team. |
| make_trigger_webhookA | Trigger a Make.com scenario by sending a POST to its webhook URL (from make_list_hooks). This FIRES the scenario for real. 'data' is an optional JSON payload sent to the webhook. |
| make_list_connectionsA | List Make.com connections (linked app accounts) for a team — id, name and app. Useful to see which integrations are connected. team_id falls back to the configured team or the only team. |
| make_test_connectionA | Verify whether a Make.com connection is still valid (Make re-checks the saved credentials against the app). Returns a 'verified' status. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/yuli-tonkin-bg/make-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server