Make

Make is a visual automation platform that chains apps into scenarios. There is no Stac app in Make: the built-in HTTP app calls POST /v1/chat/completions directly, and every run is covered by your stack's fixed monthly price.

Prerequisites#

• A key from your stack's API keys page. Create a dedicated one for Make so revoking it doesn't affect your other integrations — see Authentication.
• The id of the model you want to call — see Models. The examples below use qwen/qwen3.5-9b.

Configure the module#

1. Add the HTTP › Make a request module

   In the scenario editor, click +, search for HTTP and pick Make a request.

2. Fill in the request

   Use the values below. <STAC_API_KEY> is the key itself, after the word Bearer and a space.

   Field	Value
   Authentication type	No authentication — the key goes in the header
   URL	https://api.trystac.com/v1/chat/completions
   Method	POST
   Headers	Name Authorization, value Bearer <STAC_API_KEY>
   Body content type	application/json
   Body input method	JSON string
   Body content	The JSON below
   Parse response	Yes

   Body contentjson

   {
     "model": "qwen/qwen3.5-9b",
     "messages": [{ "role": "user", "content": "Name three primary colors." }]
   }

3. Run the module once

   Right-click the module and choose Run this module only, or click Run once. A 200 status means the key and URL are right.

4. Map the answer

   With Parse response on, the reply is at choices[1].message.content in the module's output — Make counts array items from 1. Map it into the next module (Slack, Google Sheets, email…).

Example#

The module above returns the standard chat completion body:

{
  "id": "chatcmpl-9f2a1c",
  "object": "chat.completion",
  "model": "qwen/qwen3.5-9b",
  "choices": [
    {
      "index": 0,
      "message": { "role": "assistant", "content": "Red, blue, and yellow." },
      "finish_reason": "stop"
    }
  ],
  "usage": { "prompt_tokens": 14, "completion_tokens": 7, "total_tokens": 21 }
}

To send data from an earlier module, map it into content — for example "content": "Summarize this order: {{1.description}}".

Limitations#

• No streaming. Make waits for the whole response, and a generation that runs past roughly 90 seconds fails with 503. Keep prompts focused, and turn on the module's error handling to retry 429 and 503 — see Errors & rate limits.
• model must be on your plan. A model your key's plan doesn't include returns 403 — see Models.
• Tools depend on the model. Sending tools only makes sense with a model that supports tool calling, and Make won't run the tool for you — the next modules have to read tool_calls and answer.
• Rate limits are per key. A scenario that fans out over many bundles can hit 429; a dedicated key keeps that from throttling your other apps. See Differences that matter.

Next steps#

• Chat completions — every field you can add to the body.
• Knowledge base — ground answers in your own documents with no extra modules.
• n8n — the same idea in n8n, with a native node.