UNITHdocs
Sign inarrow_forward
API

Make Integration

This guide explains how to integrate UNITH Digital Humans with Make.com using Webhooks — without writing a single line of code.

This integration allows your Digital Human to send structured user data (like names, emails, issue descriptions) to Make in real time, and optionally receive dynamic responses.

Note: Unlike with Zapier, there is currently no native UNITH integration with Make. However, full interoperability is possible using HTTP POST requests and Make’s flexible scenario engine. No developer skills are required. Perfect for business, operations, marketing and customer support teams.

We’re actively working on a native Make app as well, which will be available soon to make integration even easier.

What is Make?

Make is a visual, no-code automation platform that allows you to connect apps and services via scenarios composed of modular logic blocks. It supports custom HTTP connections, data mapping, scheduling, branching, and much more.

You build scenarios by dragging and dropping modules (like Webhooks, Gmail, Google Sheets, Slack, etc.) into a flowchart. Each module performs an action: receives data, processes it, sends a message, updates a file — whatever you need.

In the UNITH context, Make is ideal for:

  • Receiving structured user data from a Digital Human.
  • Forwarding that data to CRMs, spreadsheets, or ticketing systems.
  • Sending custom responses back to the Digital Human in real time.

Use Case

Allow a UNITH Digital Human (in OC Mode) to collect user information (e.g. name, email, issue) and send it to a Webhook hosted in Make. The scenario processes this information and returns a formatted response that the avatar can use in the conversation.

Architecture Flow

[User interacts with Digital Human] ↓ [LLM detects a trigger from prompt instructions] ↓ [Digital Human invokes a Tool → sends POST] ↓ [Make receives the POST via Webhook module] ↓ [Make processes the data (optional: trigger external services)] ↓ [Make returns a JSON payload via Webhook Response module] ↓ [Digital Human uses this payload to generate a conversational reply]

Prerequisites

Before you begin, you’ll need:

  • A UNITH Digital Human with OC Mode enabled.
  • Access to the UNITH API.
  • A Make account and access to create Webhook-based scenarios.

Step 1: Set Up Your Webhook in Make

  1. Log in to your Make.com account.
  2. Create a new Scenario.
  3. In the scenario canvas, click "+ Add module".
  4. Search for and select the "Webhooks" module.
  5. Choose "Custom webhook", then click "Add" to create a new webhook.
  6. Name your webhook (e.g., create_ticket) and click Save.
  7. Make will generate a Webhook URL like https://hook.make.com/abc123....

📌 Copy this URL — you’ll use it in the Tool definition in Step 2.

⚠️ This Webhook will wait for the tools (POST requests) from the Digital Human.

Step 2: Define a Tool for the Digital Human

Tools allow your Digital Human to make external HTTP requests using user-provided data. You’ll now define a Tool that sends a POST request to the Make webhook.

Tool JSON Example

{ "conversationSettings": { "tools": [ { "name": "Create_Ticket", "description": "Submit a support ticket with user-provided information", "parameters": [ { "name": "userName", "type": "string", "description": "User's full name" }, { "name": "userEmail", "type": "string", "description": "User's email address" }, { "name": "userIssue", "type": "string", "description": "Issue description" } ], "url": "https://hook.make.com/your-webhook-id" } ] } }

Apply the Tool Using the UNITH API

bashCopyEditcurl -X PUT \ 'https://platform-api.unith.ai/head/<HEAD_ID>/conversation-settings' \ -H 'Authorization: Bearer <YOUR_API_TOKEN>' \ -H 'Content-Type: application/json' \ -d '<YOUR_TOOL_JSON>'

Replace:

  • <HEAD_ID> with your Digital Human’s ID
  • <YOUR_API_TOKEN> with your UNITH API token
  • <YOUR_TOOL_JSON> with the JSON structure shown above

✅ Once this is applied, your Digital Human will be ready to use the Make webhook.

Step 3: Define Prompt Instructions

Your Digital Human needs clear instructions to know when to call the Tool, and what to do with the response.

Example Instruction for the Prompt:

If the user wants to open a support ticket, collect their name, email, and a description of the issue. As soon as the user provides the issue description, immediately call the Create_Ticket tool with the parameters 'userName', 'userEmail' and 'userIssue'.

Be sure the parameter names used in the prompt exactly match the ones defined in the Tool (userName, userEmail, userIssue).

Step 4: Build the Make Scenario Logic

Now it’s time to design what happens inside Make when the webhook is triggered.

A simple scenario structure:

[Webhook] ↓ [Optional: Store data, send notifications, call APIs] ↓ [HTTP Response module]

  1. After the Webhook module, add any processing step (e.g., Google Sheets, Gmail, Slack).
  2. Add the "HTTP" → "Make a response" module as the last step in your scenario.
  3. In the HTTP response (Body), return a simple JSON object:

Example Output (body):

{ "message": "Tell the user: Thank you, {{userName}}. We have received your issue: '{{userIssue}}'." }

This message will be returned to the Digital Human as structured data.

Note: The avatar won’t automatically say this message. You must tell the LLM (in the prompt) to use the message value as the reply to the user.

Step 5: Test & Validate the Flow

Once the scenario is complete:

  1. Click the "Run once" button in Make.
  2. Use the Digital Human in a live or test conversation to trigger the Tool.
  3. Make will receive the tool (POST) and execute the scenario.
  4. You can inspect the request, response, and output inside Make's Scenario Run Logs.
  5. If needed, you can re-run or debug any step.

Summary

By connecting UNITH and Make via Webhooks, you can:

  • 🚀 Trigger workflows with no code
  • 📩 Store or forward user input to third-party tools
  • 🧠 Send dynamic responses back into the conversation
  • 🎯 Empower non-technical teams to build powerful automations

Sample Use Cases

With Make + UNITH you can:

  • Save leads to Google Sheets or Airtable
  • Trigger support tickets in platforms like Zendesk or Freshdesk
  • Send email notifications or internal alerts
  • Integrate with CRMs or internal APIs
  • Log feedback or NPS scores automatically

Need Assistance?

If you need help setting this up, feel free to reach out to our team at support@unith.ai.

scheduleLast updated Jun 11, 2025