> For the complete documentation index, see [llms.txt](https://docs.snackprompt.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.snackprompt.com/bring-your-data-into-ai/how-to/get-started-with-integrations/how-to-integrate-with-make.md). # How to Integrate with Make Learn how to use the SnackPrompt AI Engine API to build powerful automations and AI-powered scenarios in Make (formerly Integromat). ### Overview Make offers several ways to integrate external APIs into your automations: | Method | Use Case | Description | | --------------------- | ---------------- | ------------------------------------------------ | | **HTTP Module** | Direct API calls | Make HTTP requests to any API | | **Webhooks** | Receive data | Trigger scenarios when external events occur | | **JSON Module** | Data processing | Parse and create JSON structures | | **OpenAI/AI Modules** | AI integration | Combine with AI models for intelligent workflows | ### Integration Architecture ``` ┌─────────────────────────────────────────────────────────┐ │ Make │ │ ┌─────────────┐ ┌─────────────┐ ┌────────────┐ │ │ │ Trigger │───▶│ HTTP Module │───▶│ Action │ │ │ └─────────────┘ └──────┬──────┘ └────────────┘ │ │ │ │ │ ┌──────▼──────┐ │ │ │ Router │ │ │ │ (optional) │ │ │ └──────┬──────┘ │ └────────────────────────────┼────────────────────────────┘ │ ▼ ┌──────────────────────────────┐ │ SnackPrompt AI Engine API │ │ /v1/kb/search or /v1/kb/chat│ └──────────────────────────────┘ ``` *** ### Method 1: HTTP Module (Recommended) Use the **HTTP > Make a request** module to call the SnackPrompt AI Engine API directly. #### Step 1: Create a New Scenario 1. Go to [make.com](https://make.com) and click **Create a new scenario** 2. Choose your trigger module (e.g., Webhook, Email, Schedule) #### Step 2: Add HTTP Module 1. Click **+** to add a new module 2. Search for **HTTP** 3. Select **Make a request** #### Step 3: Configure the Request **For Search Endpoint:**
FieldValue
URLhttps://api-integrations.snackprompt.com/v1/kb/search
MethodPOST
Body typeRaw
Content typeJSON (application/json)
**Headers:** | Name | Value | | ----------- | -------------- | | `x-api-key` | `YOUR_API_KEY` | **Request content (JSON):** ```json { "query": "{{1.query}}", "filters": { "tenant_id": "YOUR_TENANT_ID" }, "limit": 5 } ``` Replace `{{1.query}}` with the appropriate variable from your trigger module. #### Step 4: Parse the Response Check **Parse response** to automatically parse the JSON response into usable variables. #### Step 5: Use the Results The API returns a list of relevant documents in `items[]`. You can use them in subsequent modules: * `{{2.items[].payload.original_text}}` - The document content * `{{2.items[].score}}` - Relevance score * `{{2.items[].payload.snack_item_id}}` - Document ID *** ### Method 2: Webhooks (Receive Requests) Use when you want to **receive requests** and respond with knowledge base information. #### Step 1: Create Webhook Trigger 1. Create a new scenario 2. Add **Webhooks > Custom webhook** as the trigger 3. Click **Add** to create a new webhook 4. Copy the webhook URL provided #### Step 2: Add HTTP Module Configure as shown in Method 1, using the webhook data: **Request content:** ```json { "query": "{{1.query}}", "filters": { "tenant_id": "YOUR_TENANT_ID" }, "limit": 5 } ``` #### Step 3: Return Response Add **Webhooks > Webhook response** module at the end: | Field | Value | | -------------- | ---------------------------------- | | Status | `200` | | Body | `{{2.body}}` or formatted response | | Custom headers | `Content-Type: application/json` | *** ### Method 3: Chat Endpoint for Complete Responses Use the `/v1/kb/chat` endpoint when you want the **API to handle all the RAG** and return a ready response. #### Configuration
FieldValue
URLhttps://api-integrations.snackprompt.com/v1/kb/chat
MethodPOST
Body typeRaw
Content typeJSON (application/json)
**Headers:** | Name | Value | | ----------- | -------------- | | `x-api-key` | `YOUR_API_KEY` | **Request content:** ```json { "query": "{{1.message}}", "filters": { "tenant_id": "YOUR_TENANT_ID", "tag_names": ["Support", "FAQ"] } } ``` #### Response The API returns: * `answer`: AI-generated response ready to use * `sources`: Sources used to generate the response Access with `{{2.answer}}` and `{{2.sources}}`. *** ### Method 4: With AI Modules (RAG Pipeline) Combine the search API with Make's AI modules for custom RAG workflows. #### RAG Scenario Structure ``` [Trigger] → [HTTP: Search API] → [Text Aggregator] → [OpenAI: Create Completion] → [Action] ``` #### Step 1: Search for Context Use HTTP module to call `/v1/kb/search` as shown in Method 1. #### Step 2: Format Context with Iterator + Aggregator **Add Iterator module:** * Array: `{{2.items}}` **Add Text Aggregator module:** * Source module: Iterator * Text: `[{{3.i}}] {{3.payload.original_text}}` * Row separator: `\n\n` #### Step 3: Generate Response with OpenAI **Add OpenAI > Create a Completion module:** | Field | Value | | -------- | -------------------------- | | Model | `gpt-4` or `gpt-3.5-turbo` | | Messages | System + User messages | **System Message:** ``` You are a helpful assistant. Answer the user's question based ONLY on the following context. Context: {{4.text}} If the context doesn't contain relevant information, say "I don't have information about that." ``` **User Message:** ``` {{1.query}} ``` *** ### Practical Use Cases #### 1. Support Chatbot ``` [Webhook] → [HTTP: /chat] → [Webhook Response] ``` Simple chatbot that answers questions using your knowledge base. #### 2. Email Auto-Responder ``` [Email: Watch] → [HTTP: /chat] → [Email: Send] ``` Automatically respond to customer emails with relevant information. #### 3. Slack Integration ``` [Slack: Watch Channel] → [HTTP: /search] → [Slack: Post Message] ``` Answer questions posted in a Slack channel. #### 4. Multi-Source RAG with Router ``` [Webhook] → [Router] │ ├─[Filter: Sales]──→ [HTTP: tag=Sales] ───┐ │ │ └─[Filter: Support]→ [HTTP: tag=Support]──┼→ [Array Aggregator] → [OpenAI] → [Response] ``` Route queries to different knowledge bases based on topic. #### 5. Scheduled Knowledge Digest ``` [Schedule: Weekly] → [HTTP: /search trending] → [Iterator] → [Notion: Create Page] ``` Generate weekly reports from your knowledge base. #### 6. Form Response with Fallback ``` [Typeform: Watch] → [HTTP: /search] → [Router] │ ├─[Filter: results > 0]→ [Email: Send Answer] │ └─[Filter: results = 0]→ [Zendesk: Create Ticket] ``` *** ### Configuration Tips #### 1. Store Credentials Securely Use Make's **Connections** or **Data Stores** for API keys: 1. Go to **Connections** in the left menu 2. Create a new **HTTP Basic Auth** or **API Key** connection 3. Use the connection in your HTTP modules #### 2. Use Variables and Data Stores Store `tenant_id` and other config in a Data Store: ```json { "query": "{{1.query}}", "filters": { "tenant_id": "{{datastore.config.tenant_id}}" } } ``` #### 3. Handle Errors with Error Handler Right-click the HTTP module and add an **Error Handler**: ``` [HTTP Module] ──error──→ [Router] │ ├─[Filter: 4xx]→ [Slack: Notify] │ └─[Filter: 5xx]→ [Resume] → [Sleep] → [Retry] ``` #### 4. Limit Results for Efficiency Start with fewer results: ```json { "limit": 3 } ``` #### 5. Use Filters for Context Direct searches with tags: ```json { "filters": { "tenant_id": "...", "tag_names": ["{{1.detected_category}}"] } } ``` #### 6. Enable Parse Response Always check **Parse response** in the HTTP module to easily access response fields. *** ### Complete Example: Customer Support Automation #### Scenario Overview 1. Customer submits question via form 2. Scenario searches knowledge base 3. If answer found, sends automated reply 4. If not found, creates support ticket #### Step-by-Step Setup **Module 1: Typeform - Watch Responses** * Connection: Your Typeform account * Form: Select your support form **Module 2: HTTP - Make a request** * URL: `https://api-integrations.snackprompt.com/v1/kb/chat` * Method: `POST` * Headers: `x-api-key: YOUR_API_KEY` * Body type: `Raw` * Content type: `JSON (application/json)` * Request content: ```json { "query": "{{1.answers[].text}}", "filters": { "tenant_id": "YOUR_TENANT_ID", "tag_names": ["Support"] } } ``` * Parse response: `Yes` **Module 3: Router** * Route 1: `{{length(2.sources)}} > 0` (has results) * Route 2: `{{length(2.sources)}} = 0` (no results) **Module 4A: Gmail - Send an Email** (Route 1) * To: `{{1.answers[email].email}}` * Subject: `Re: Your question` * Content: `{{2.answer}}` **Module 4B: Zendesk - Create Ticket** (Route 2) * Subject: `Support Request: {{1.answers[subject].text}}` * Description: `{{1.answers[question].text}}` *** ### Working with Arrays #### Iterating Over Results Use **Iterator** to process each result: ``` [HTTP: /search] → [Iterator] → [Your Module] │ Array: {{2.items}} ``` Inside the iterator, access: * `{{3.payload.original_text}}` * `{{3.score}}` * `{{3.payload.snack_item_id}}` #### Aggregating Results Use **Text Aggregator** to combine results: | Field | Value | | ------------- | ------------------------------- | | Source module | Iterator | | Text | `- {{3.payload.original_text}}` | | Row separator | `\n` | Result: A formatted list of all documents. #### Array Functions Useful functions for working with results: | Function | Example | Description | | ---------- | -------------------------- | ------------------- | | `first()` | `{{first(2.items)}}` | Get first result | | `last()` | `{{last(2.items)}}` | Get last result | | `length()` | `{{length(2.items)}}` | Count results | | `slice()` | `{{slice(2.items; 0; 3)}}` | Get first 3 results | *** ### Troubleshooting #### Error: "tenant\_id is required" Make sure `tenant_id` is inside the `filters` object: ```json // ❌ Wrong { "query": "...", "tenant_id": "..." } // ✅ Correct { "query": "...", "filters": { "tenant_id": "..." } } ``` #### Error: "Invalid JSON" 1. Verify JSON syntax in the request content 2. Check for unescaped special characters in variables 3. Use the **JSON > Create JSON** module to build complex payloads safely #### Empty Response / No Items 1. Verify your `tenant_id` is correct 2. Check if the query matches content in your knowledge base 3. Remove `tag_names` filter to search all content 4. Increase the `limit` parameter #### Cannot Access Response Fields 1. Ensure **Parse response** is checked 2. Verify the response structure by checking execution history 3. Use `{{2.body}}` to see the raw response #### Rate Limiting (429 Error) 1. Add a **Sleep** module between API calls 2. Use Make's built-in **Rate limiting** in scenario settings 3. Consider using **Queue** for high-volume scenarios #### Timeout Errors 1. Increase the **Timeout** setting in the HTTP module 2. Check if your query is too complex 3. Reduce the `limit` parameter *** ### Advanced Patterns #### Caching with Data Stores Cache frequent queries to reduce API calls: ``` [Trigger] → [Data Store: Search] → [Router] │ ├─[Found]→ [Use Cache] │ └─[Not Found]→ [HTTP: /search] → [Data Store: Add] → [Continue] ``` #### Retry Pattern Automatic retry on failure: ``` [HTTP Module] ──error──→ [Tools: Sleep] → [Tools: Increment] → [Router] │ ├─[retries < 3]→ [Resume] │ └─[retries >= 3]→ [Error Notification] ``` #### Batch Processing Process multiple queries efficiently: ``` [Trigger with Array] → [Iterator] → [HTTP: /search] → [Array Aggregator] → [Next Module] ``` *** ### Related * [Endpoints Reference](/bring-your-data-into-ai/reference/endpoints.md) * [Available Filters](/bring-your-data-into-ai/reference/filters.md) * [Error Hand](/bring-your-data-into-ai/how-to/how-to-handle-errors.md) ### External Resources * [Make HTTP Module Documentation](https://www.make.com/en/help/app/http) * [Make Webhooks Guide](https://www.make.com/en/help/app/webhooks) * [Make Error Handling](https://www.make.com/en/help/errors/error-handling) * [Make JSON Module](https://www.make.com/en/help/app/json) --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://docs.snackprompt.com/bring-your-data-into-ai/how-to/get-started-with-integrations/how-to-integrate-with-make.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.