Make integration for RAG workflows

Connect Biel.ai to Make (formerly Integromat) to query your indexed documentation from any scenario. Use it to build support assistants, Slack bots, email automation, or any workflow that needs answers from your docs.

info

This integration requires a Business or Enterprise plan for API access.

Prerequisites

• A Biel.ai account with documentation indexed.
• A Business or Enterprise plan.
• A Make account.
• Your project slug from the Biel.ai dashboard.
• An API key from Team → API keys, with Project access set to your project and chats_create permission.

Setup

1. Create a new scenario

Create a new scenario with the following modules:

1. Trigger module (e.g., Webhook, Custom webhook, Slack)
2. HTTP module to query Biel.ai
3. Response module to return results

2. Configure the HTTP module

Add an HTTP > Make a request module with these settings:

URL: https://app.biel.ai/api/v2/projects/YOUR_PROJECT_SLUG/chats/

Method: POST

Headers:

Key	Value
Content-Type	application/json
Authorization	Api-Key YOUR_API_KEY

Replace YOUR_API_KEY with your API key from your Biel.ai team settings.

Body:

{
  "message": "{{1.question}}"
}

Replace {{1.question}} with the appropriate variable from your trigger module. Replace YOUR_PROJECT_SLUG with your project slug from the Biel.ai dashboard.

For the full API reference, see Biel.ai REST API.

3. Process the response

The chat API returns a structured response:

{
  "user_message_id": "message-id",
  "ai_message": {
    "from": "ai",
    "message": "AI-generated answer based on your documentation...",
    "messageId": "ai-message-id",
    "isPartial": false,
    "timestamp": "2024-01-01T00:00:00Z",
    "sources": [
      {
        "title": "Source page title",
        "url": "https://docs.example.com/source"
      }
    ]
  },
  "chat_uuid": "unique-conversation-id",
  "restore_token": "opaque-conversation-token"
}

Use the returned data in subsequent modules:

• {{2.ai_message.message}} - The AI-generated answer
• {{2.ai_message.sources}} - Array of source documents with citations
• {{2.chat_uuid}} - Conversation ID for multi-turn chats

4. Continue the conversation (optional)

Retain both chat_uuid and restore_token from the response. For each subsequent request to the same project endpoint, include chat_uuid in the body and add this header:

Key	Value
X-Chat-Restore-Token	{{2.restore_token}}

{
  "message": "{{1.question}}",
  "chat_uuid": "{{2.chat_uuid}}"
}