Make Integration API Reference

The AssistLoop API is specifically designed to integrate seamlessly with Make (formerly Integromat), enabling you to build powerful automation scenarios with visual workflows. Our API follows REST principles with predictable resource-oriented URLs, accepts JSON-encoded request bodies, returns JSON-encoded responses, and uses standard HTTP response codes and OAuth 2.0 authentication.

What You Can Build

With the AssistLoop Make integration, you can:

Trigger Scenarios when new conversations start or when AI requests human assistance
Automatically respond to customer conversations from other apps
Close conversations as part of your scenario automation
Query conversation history to feed data into your other tools
Build custom scenarios connecting AssistLoop with 1,500+ apps on Make

Base URL

https://api.assistloop.ai/v1

Authentication

The AssistLoop Make integration uses OAuth 2.0 for secure authentication. When you connect AssistLoop to Make, you'll be prompted to authorize the connection through a secure OAuth flow. Make will automatically handle token management, including automatic token refresh when needed.

For Make Users

You don't need to manually manage tokens. Simply click "Add" next to the Connection field in any AssistLoop module and follow the authorization prompts.

For Developers

If you're building custom Make modules or testing the API directly, all requests must include your access token in the Authorization header:

Authorization: Bearer YOUR_ACCESS_TOKEN

All API requests must be made over HTTPS. Calls made over plain HTTP will fail.

Example Request with Authentication

curl https://api.assistloop.ai/v1/oauth/me \
  -H "Authorization: Bearer YOUR_ACCESS_TOKEN"

OAuth 2.0 Authorization Flow

AssistLoop uses OAuth 2.0 to securely connect with Make. This section describes the technical details of how Make obtains and refreshes access tokens when you connect your AssistLoop account.

Most Make users won't need to understand these technical details - Make handles the OAuth flow automatically. This section is primarily for developers building or debugging Make integrations.

OAuth 2.0 Endpoints

Endpoint	URL	Purpose
Authorization	`POST /api/v1/oauth/authorize`	Obtain authorization code
Token	`POST /oauth/token`	Exchange code for token or refresh access token

The /oauth/token endpoint is at the root level (not under /api/v1).

Available Scopes

Scope	Description
`read:conversations`	Read conversations and conversation details (default)
`read:messages`	Read messages within conversations (default)
`manage:webhooks`	Create and manage webhook subscriptions (default)
`write:conversations`	Update conversation status and response mode

Default Scopes: If no scopes are specified, the token will have read:conversations, read:messages, and manage:webhooks scopes.

Make OAuth Configuration

When setting up your Make integration, use the following OAuth configuration:

OAuth Configuration Parameters

Parameter	Value
Authorization URL	`https://api.assistloop.ai/v1/oauth/authorize`
Token URL	`https://api.assistloop.ai/oauth/token`
Scope	`read:conversations read:messages manage:webhooks write:conversations`
Client ID	(Provided in AssistLoop Dashboard)
Client Secret	(Provided in AssistLoop Dashboard)
Redirect URI	`https://www.integromat.com/oauth/cb/app`

Make Connection Settings

Connection Name: AssistLoop OAuth
Authorization Method: OAuth 2.0
Grant Type: Authorization Code
Show advanced settings: Yes
Authorization URL Structure: Custom
Token Request Method: POST (JSON)
Scope Separator: Space
Automatically Refresh: Yes

Test API Call

Configure Make's test API call to verify authentication:

URL: https://api.assistloop.ai/v1/oauth/me
Method: GET
Expected Response: JSON with uuid, name, and email fields

Errors

AssistLoop uses conventional HTTP response codes to indicate the success or failure of an API request. In general: Codes in the 2xx range indicate success. Codes in the 4xx range indicate an error that failed given the information provided (e.g., a required parameter was omitted, a resource was not found, etc.). Codes in the 5xx range indicate an error with AssistLoop's servers (these are rare).

HTTP Status Code Summary

Code	Status	Description
200	OK	Everything worked as expected.
400	Bad Request	The request was unacceptable, often due to missing a required parameter.
401	Unauthorized	No valid API token provided.
403	Forbidden	The API token doesn't have permissions to perform the request.
404	Not Found	The requested resource doesn't exist.
422	Validation Error	The request parameters failed validation.
429	Too Many Requests	Too many requests hit the API too quickly. Please retry with exponential backoff.
500	Internal Server Error	Something went wrong on AssistLoop's end.

Error Response Format

{
    "error": "Conversation not found",
    "message": "The requested conversation could not be found or you don't have access to it."
}

OAuth Endpoints

Get Current User

Retrieves information about the currently authenticated user. Use this endpoint to verify your authentication token and get details about the associated account.

Endpoint: GET /oauth/me

Example Request

curl https://api.assistloop.ai/v1/oauth/me \
  -H "Authorization: Bearer YOUR_ACCESS_TOKEN"

Response

{
    "uuid": "usr_abc123",
    "name": "John Doe",
    "email": "[email protected]"
}

Webhook Endpoints (Make Instant Triggers)

Webhooks power Make instant triggers for your AssistLoop account. When you set up an instant trigger in Make (like "New Conversation" or "Human Handoff Required"), Make automatically subscribes to the appropriate webhook endpoint. When the event occurs in AssistLoop, we instantly notify Make to trigger your scenario.

Creates a new webhook subscription for Make instant triggers. When you enable an instant trigger in Make, Make automatically calls this endpoint to subscribe to events. When the event occurs in AssistLoop, we send the data to Make to trigger your scenario.

Endpoint: POST /webhooks/subscribe

For Make Users: This happens automatically when you set up an instant trigger. You don't need to call this endpoint manually.

Request Parameters

Parameter	Type	Required	Description
organization_uuid	string	Yes	The UUID of your organization (workspace).
agent_uuid	string	Yes	The UUID of the agent to monitor for events.
target_url	string	Yes	The URL where webhook payloads will be sent. Must be a valid HTTPS URL.
event	string	Yes	The type of event to subscribe to. One of: `new_conversation`, `human_handoff_required`.
platform	string	Yes	Must be `make` for Make integrations.
external_id	string	No	Optional external identifier for your integration (max 255 characters).

Example Request

curl -X POST https://api.assistloop.ai/v1/webhooks/subscribe \
  -H "Authorization: Bearer YOUR_ACCESS_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "organization_uuid": "org_xyz789",
    "agent_uuid": "agt_abc123",
    "target_url": "https://hook.integromat.com/abcd1234efgh5678",
    "event": "new_conversation",
    "platform": "make"
  }'

Response

{
    "data": {
        "id": "whk_xyz789",
        "secret": "whsec_a1b2c3d4e5f6g7h8i9j0k1l2m3n4o5p6"
    },
    "message": "Webhook subscribed successfully"
}

Important

The secret field is returned only once during subscription. Store it securely to verify webhook signatures.

Webhook Event Types (Make Instant Triggers)

New Conversation

Event: new_conversation

Triggers your scenario when a customer starts a new conversation with your AI agent. Perfect for logging conversations, sending notifications, or creating tickets.

Human Handoff Required

Event: human_handoff_required

Triggers your scenario when the AI determines human assistance is needed. Use this to notify your support team, create urgent tickets, or escalate to your help desk system.

Webhook Payload Examples

New Conversation Event

{
    "id": "550e8400-e29b-41d4-a716-446655440000",
    "event": "new_conversation",
    "timestamp": "2024-11-12T10:30:00Z",
    "data": {
        "conversation": {
            "uuid": "550e8400-e29b-41d4-a716-446655440000",
            "status": true,
            "response_mode": "ai",
            "created_at": "2024-11-12T10:30:00Z"
        },
        "agent": {
            "uuid": "agt_abc123",
            "name": "Customer Support Agent"
        },
        "chat_user": {
            "name": "Jane Smith",
            "email": "[email protected]"
        },
        "organization": {
            "uuid": "org_xyz789",
            "name": "Acme Corporation"
        },
        "initial_message": {
            "uuid": "msg_abc123",
            "content": "Hi, I need help with my order",
            "type": "guest",
            "created_at": "2024-11-12T10:30:00Z"
        }
    }
}

Human Handoff Required Event

{
    "id": "550e8400-e29b-41d4-a716-446655440000",
    "event": "human_handoff_required",
    "timestamp": "2024-11-12T10:35:00Z",
    "data": {
        "conversation": {
            "uuid": "550e8400-e29b-41d4-a716-446655440000",
            "status": true,
            "response_mode": "human",
            "created_at": "2024-11-12T10:30:00Z"
        },
        "agent": {
            "uuid": "agt_abc123",
            "name": "Customer Support Agent"
        },
        "chat_user": {
            "name": "Jane Smith",
            "email": "[email protected]"
        },
        "organization": {
            "uuid": "org_xyz789",
            "name": "Acme Corporation"
        },
        "last_message": {
            "uuid": "msg_def456",
            "content": "I need to speak with a human agent",
            "type": "guest",
            "created_at": "2024-11-12T10:35:00Z"
        }
    }
}

Unsubscribe from Webhooks

Removes a webhook subscription. Make automatically calls this endpoint when you turn off or delete an instant trigger.

Endpoint: DELETE /webhooks/unsubscribe/{uuid}

For Make Users: This happens automatically when you disable or delete an instant trigger. You don't need to call this endpoint manually.

Path Parameters

Parameter	Type	Required	Description
uuid	string	Yes	The UUID of the webhook subscription.

Example Request

curl -X DELETE https://api.assistloop.ai/v1/webhooks/unsubscribe/whk_xyz789 \
  -H "Authorization: Bearer YOUR_ACCESS_TOKEN"

Response

{
    "data": null,
    "message": "Webhook unsubscribed successfully"
}

Remote Procedures (Make Dropdowns)

Remote Procedure Calls (RPCs) power the dropdown selectors in Make modules, allowing you to dynamically select workspaces, agents, and conversations.

Get Workspaces

Retrieve a list of workspaces (organizations) you are a member of. Used in Make dropdown selectors.

Endpoint: GET /my/organizations

Example Request

curl https://api.assistloop.ai/v1/my/organizations \
  -H "Authorization: Bearer YOUR_ACCESS_TOKEN"

Response

[
    {
        "organization_uuid": "org_abc123xyz",
        "id": "org_abc123xyz",
        "name": "Acme Corp"
    },
    {
        "organization_uuid": "org_def456uvw",
        "id": "org_def456uvw",
        "name": "Tech Startup Inc"
    }
]

Get Agents

Retrieve a list of AI agents in a specific workspace. Used in Make dropdown selectors.

Endpoint: GET /organizations/{organization_uuid}/agents

Path Parameters

Parameter	Type	Description
organization_uuid	string	UUID of the workspace

Example Request

curl https://api.assistloop.ai/v1/organizations/org_abc123xyz/agents \
  -H "Authorization: Bearer YOUR_ACCESS_TOKEN"

Response

[
    {
        "agent_uuid": "agent_abc123",
        "id": "agent_abc123",
        "name": "Support Bot"
    },
    {
        "agent_uuid": "agent_def456",
        "id": "agent_def456",
        "name": "Sales Assistant"
    }
]

Get Conversations

Retrieve a list of conversations for a specific agent. Used in Make dropdown selectors.

Endpoint: GET /automation/agents/{agent_uuid}/conversations/list

Path Parameters

Parameter	Type	Description
agent_uuid	string	UUID of the agent

Query Parameters

Parameter	Type	Required	Description
per_page	number	No	Number of results (default: 100)

Example Request

curl "https://api.assistloop.ai/v1/automation/agents/agent_abc123/conversations/list?per_page=100" \
  -H "Authorization: Bearer YOUR_ACCESS_TOKEN"

Response

[
    {
        "id": "conv_abc123xyz",
        "uuid": "conv_abc123xyz",
        "session_id": "sess_xyz789",
        "response_mode": "ai",
        "status": true,
        "last_message_seen": false,
        "chat_user": {
            "name": "John Doe",
            "email": "[email protected]",
            "country_code": "+1",
            "chat_user_id": "user_12345"
        }
    }
]

Conversation Endpoints (Make Actions)

List Conversations

Query and filter conversations for a specific agent. Use this in Make scenarios to find conversations matching specific criteria.

Endpoint: GET /automation/agents/{agentUuid}/conversations/list

Make Use Cases

Find open conversations to send daily summaries to Slack
Check for conversations in "human" mode to assign to support agents
Export conversation lists to Google Sheets for reporting

Path Parameters

Parameter	Type	Required	Description
agentUuid	string	Yes	The UUID of the agent.

Query Parameters

Parameter	Type	Required	Description
status	integer	No	Filter by conversation status. `0` for closed, `1` for active/open.
response_mode	string	No	Filter by response mode. One of: `ai`, `human`.
created_from	string	No	Filter conversations created after this date (ISO 8601 format).
created_to	string	No	Filter conversations created before this date (ISO 8601 format).
per_page	integer	No	Number of conversations per page (min: 10, max: 100, default: 20).

Example Request

curl "https://api.assistloop.ai/v1/automation/agents/agt_abc123/conversations/list?status=1&response_mode=ai&per_page=20" \
  -H "Authorization: Bearer YOUR_ACCESS_TOKEN"

Response

The endpoint returns a direct array of conversation objects (not wrapped in a data field).

[
    {
        "id": "550e8400-e29b-41d4-a716-446655440000",
        "uuid": "550e8400-e29b-41d4-a716-446655440000",
        "session_id": "sess_abc123",
        "response_mode": "ai",
        "status": true,
        "last_message_seen": false,
        "deleted_at": null,
        "chat_user": {
            "name": "Jane Smith",
            "email": "[email protected]",
            "country_code": "US",
            "chat_user_id": "chu_xyz123"
        }
    }
]

Field Descriptions

status: Boolean - true for active/open, false for closed
response_mode: String - Either ai (AI responding) or human (human agent responding)
last_message_seen: Boolean - Whether the last message has been seen
chat_user: Object containing customer information

Get Recent Conversation Samples

Provides sample data when you're setting up a Make instant trigger. When you click "Run once" to test the webhook in Make, this endpoint returns recent conversations so you can see what data your scenario will receive.

Endpoint: GET /automation/agents/{agentUuid}/conversations/recent-sample

For Make Users: This runs automatically when you test your instant trigger during scenario setup.

Path Parameters

Parameter	Type	Required	Description
agentUuid	string	Yes	The UUID of the agent.

Query Parameters

Parameter	Type	Required	Description
limit	integer	No	Number of samples to return (default: 5, max: 10).
type	string	No	Sample type. One of: `new_conversation`, `human_handoff_required` (default: `new_conversation`).

Example Request

curl "https://api.assistloop.ai/v1/automation/agents/agt_abc123/conversations/recent-sample?limit=3&type=new_conversation" \
  -H "Authorization: Bearer YOUR_ACCESS_TOKEN"

Response

Returns an array of webhook payload samples in the same format as actual webhook deliveries.

[
    {
        "id": "550e8400-e29b-41d4-a716-446655440000",
        "event": "new_conversation",
        "timestamp": "2024-11-12T10:30:00Z",
        "data": {
            "conversation": {
                "uuid": "550e8400-e29b-41d4-a716-446655440000",
                "status": true,
                "response_mode": "ai",
                "created_at": "2024-11-12T10:30:00Z"
            },
            "agent": {
                "uuid": "agt_abc123",
                "name": "Customer Support Agent"
            },
            "chat_user": {
                "name": "Sample Customer",
                "email": "[email protected]"
            },
            "organization": {
                "uuid": "org_xyz789",
                "name": "Acme Corporation"
            },
            "initial_message": {
                "uuid": "msg_sample123",
                "content": "Hello, I have a question",
                "type": "guest",
                "created_at": "2024-11-12T10:30:00Z"
            }
        }
    }
]

When type=human_handoff_required, the response format changes slightly - initial_message becomes last_message and the event field is set to human_handoff_required.

Close Conversation

Use this Make action to automatically close conversations as part of your workflow. Great for marking conversations as complete after creating a help desk ticket, logging to your CRM, or after a certain amount of time.

Endpoint: POST /automation/conversations/{conversationUuid}/close

Rate Limit

10 requests per minute

Make Use Cases

Close conversations after creating a Zendesk ticket
Auto-close conversations that have been idle for 24 hours
Mark conversations as complete after successful payment in Stripe

Path Parameters

Parameter	Type	Required	Description
conversationUuid	string	Yes	The UUID of the conversation.

Request Parameters

Parameter	Type	Required	Description
platform	string	Yes	Must be `make` for Make integrations.

Example Request

curl -X POST https://api.assistloop.ai/v1/automation/conversations/550e8400-e29b-41d4-a716-446655440000/close \
  -H "Authorization: Bearer YOUR_ACCESS_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "platform": "make"
  }'

Response

{
    "status": "conversation_closed",
    "platform": "make"
}

Message Endpoints (Make Actions)

Find Messages

Use this Make action to retrieve the full message history from a conversation. Perfect for including context when creating support tickets, logging conversations, or analyzing customer interactions.

Endpoint: GET /automation/conversations/{conversationUuid}/messages/find

Make Use Cases

Fetch conversation history when creating a Zendesk ticket
Log full conversation to Airtable or Google Sheets
Send conversation transcript via email when human handoff is requested

Path Parameters

Parameter	Type	Required	Description
conversationUuid	string	Yes	The UUID of the conversation.

Query Parameters

Parameter	Type	Required	Description
created_from	string	No	Filter messages created after this date (ISO 8601 format).
created_to	string	No	Filter messages created before this date (ISO 8601 format).
per_page	integer	No	Number of messages per page (min: 10, max: 100, default: 20).

Example Request

curl "https://api.assistloop.ai/v1/automation/conversations/550e8400-e29b-41d4-a716-446655440000/messages/find?per_page=20" \
  -H "Authorization: Bearer YOUR_ACCESS_TOKEN"

Response

Returns a direct array of message objects, ordered by created_at descending (newest first).

[
    {
        "id": "msg_abc124",
        "uuid": "660e8400-e29b-41d4-a716-446655440002",
        "content": "I'd be happy to help you with your order. Could you provide your order number?",
        "type": "agent",
        "is_read": null,
        "status": null,
        "responder_type": "ai",
        "emailed_at": null,
        "email_seen": false,
        "email_opened_at": null,
        "via": "web",
        "created_at": "2024-11-12T10:30:15.000000Z",
        "user_feedback": null,
        "sender_name": "Customer Support Agent",
        "attachments": []
    },
    {
        "id": "msg_abc123",
        "uuid": "660e8400-e29b-41d4-a716-446655440001",
        "content": "I need help with my order",
        "type": "guest",
        "is_read": null,
        "status": null,
        "responder_type": null,
        "emailed_at": null,
        "email_seen": false,
        "email_opened_at": null,
        "via": "web",
        "created_at": "2024-11-12T10:30:00.000000Z",
        "user_feedback": null,
        "sender_name": "Jane Smith",
        "attachments": []
    }
]

Field Descriptions

type: Message type - guest (customer), agent (AI or human), or platform_action (system messages - excluded from results)
responder_type: Who/what responded - ai, human_assistant, or null for guest messages
sender_name: Display name of the sender (automatically computed)
attachments: Array of file attachments associated with the message
user_feedback: Customer feedback on AI response - positive, negative, not-allowed, or null

Reply to Conversation

Use this Make action to automatically send messages to customers in response to events in other apps. Messages are sent as if from a human support agent.

Endpoint: POST /automation/conversations/{conversationUuid}/reply

Make Use Cases

Send a message when a help desk ticket is created: "We've created ticket #12345"
Notify customers when their order ships via your e-commerce platform
Reply with order status when customer asks about their order
Send proactive messages based on events in your CRM

Path Parameters

Parameter	Type	Required	Description
conversationUuid	string	Yes	The UUID of the conversation.

Request Parameters

Parameter	Type	Required	Description
content	string	Yes	The message content to send.

Example Request

curl -X POST https://api.assistloop.ai/v1/automation/conversations/550e8400-e29b-41d4-a716-446655440000/reply \
  -H "Authorization: Bearer YOUR_ACCESS_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "content": "Thank you for contacting us. A support agent will be with you shortly."
  }'

Response

{
    "status": "sent",
    "message_id": "msg_xyz789"
}

Important Notes

The message is sent as type agent with responder_type human_assistant
The conversation's response_mode is automatically switched to human
The message is broadcast in real-time to the customer
The authenticated user is recorded as the human responder

Common Make Scenario Examples

Scenario 1: Slack Notification for New Conversations

Goal

Get instant Slack notifications when customers start conversations with your AI agent.

Instant trigger that fires when a customer starts a new conversation.

Configure the Slack module to post details to your desired channel.

Configuration

Map {{data.chat_user.name}} to include customer name
Map {{data.initial_message.content}} to show first message
Include conversation URL: https://app.assistloop.ai/conversations/{{data.conversation.uuid}}

Benefits

Keep your team informed in real-time
No customer conversation goes unnoticed
Quick visibility into conversation volume

Scenario 2: Create Zendesk Ticket on Human Handoff

Goal

Automatically create a Zendesk ticket with full conversation context when AI needs human help.

Instant trigger that starts the scenario.

Retrieve conversation history for context.

Process each message and combine them into a transcript.

Use aggregated conversation history in ticket description, add customer email as requester, and set priority to high.

Message: "Thanks! We've created ticket #{{ticket_id}} and a team member will help you shortly."

Close the conversation to prevent additional AI responses.

Benefits

Zero delay between AI handoff and ticket creation
Full conversation context automatically included
Customer gets immediate acknowledgment
Conversation automatically closed to prevent AI responses

Scenario 3: Daily Conversation Summary

Goal

Get a daily digest of conversation metrics delivered to your team's Slack channel.

Start the workflow on a schedule.

Filter with created_from = yesterday's date.

Calculate metrics like totals, open vs closed, and AI vs human counts.

Share the summarized metrics with the team.

Example Slack Message

📊 Daily AssistLoop Report - November 12, 2024

✅ Total Conversations: 47
🤖 Handled by AI: 39 (83%)
👤 Required Human Help: 8 (17%)
📭 Currently Open: 5
✓ Closed: 42

🔗 View open conversations in dashboard

Scenario 4: Auto-Close Inactive Conversations

Goal

Keep your conversation list clean by automatically closing conversations that haven't had activity for 24 hours.

Kick off the clean-up scenario.

Filter for Status = Open (1).

Process each conversation and continue only if updated_at is more than 24 hours ago.

Message: "We haven't heard back from you. I'm marking this conversation as complete. Feel free to start a new conversation anytime!"

Close the inactive conversation.

Pro Tips

Adjust the time threshold based on your business hours
Customize the goodbye message based on conversation context
Add error handlers to prevent the scenario from failing on individual errors

Rate Limiting

To ensure reliable service for all Make users, we implement rate limits on API endpoints:

General Endpoints

60 requests per minute per connected account

Close Conversation

10 requests per minute

Webhook Deliveries

Automatic retries up to 3 times with exponential backoff

For Make Users: These rate limits are generous for typical scenario usage. If you hit a rate limit, Make will receive a 429 Too Many Requests response. Consider adding Sleep modules between API calls when processing many items.

429 Error Response

{
    "message": "Too Many Attempts.",
    "retry_after": 60
}

Best Practices for Make Scenarios

Designing Efficient Scenarios

Use Instant Triggers When Possible

Instant triggers (webhooks) are faster and more reliable than polling
They're more efficient and don't consume as many operations
Set up "New Conversation" and "Human Handoff" as instant triggers

Handle Multiple Results Properly

Use Iterator modules when processing multiple conversations or messages
Add error handlers to prevent the entire scenario from failing
Use filters to process only relevant items

Optimize Data Mapping

Map only the fields you actually need
Use Make's built-in functions for date/text formatting
Store frequently used values in variables or data stores

Error Handling in Make

Add Error Handlers

Right-click any module → "Add error handler"
Choose "Rollback" for critical operations
Choose "Ignore" for non-critical operations
Use "Resume" + notification for logging errors

Common Error Patterns

Error Handler → Slack notification → Stop
Error Handler → Email → Rollback → Stop
Error Handler → Log to Google Sheets → Resume

Optimizing API Calls

Reduce Unnecessary Calls

Don't fetch full message history if you only need the latest message
Use pagination wisely - start with 10-20 items
Filter conversations by date to avoid fetching old data
Cache frequently accessed data using Data Stores

Rate Limit Considerations

Add 1-2 second Sleep modules between Close Conversation actions
For bulk operations, use scheduling to spread calls over time
Monitor your scenario execution history for 429 errors

Data Handling Best Practices

Understanding Field Types

Field Type	Format	Example	Notes
Conversation Status	Boolean	`true` = open, `false` = closed	Use `1` in filters, not `true`
Response Mode	String	`ai` or `human`	Use for routing logic
Message Types	String	`guest`, `agent`	Filter customer vs AI messages
Timestamps	ISO 8601	`2024-11-12T10:30:00Z`	Use formatDate() function

Handling Missing Data

Customer email and name may be null for anonymous conversations

Use Make's if() function to provide defaults:

{{if(data.chat_user.email; data.chat_user.email; "No email provided")}}

Check for empty arrays before iterating (attachments, messages)
Use Error Handlers to catch missing required fields

Working with Message Content

Messages may contain line breaks - use replace() to clean up
Check attachments array length before accessing
User feedback field is usually null (only set if customer rated the response)
Use Text Parser modules for extracting specific information

Security Recommendations

Protect Customer Data

Don't log customer messages to public channels
Be careful when sending conversation data to third-party apps
Use Make's data encryption features for sensitive information
Regularly review which apps have access to your AssistLoop data

OAuth Token Management

Never share your AssistLoop connection with others
Use separate connections for development and production scenarios
Reconnect if you suspect token compromise
Make handles token refresh automatically - don't try to manually refresh

Webhook Security

Make automatically validates webhook signatures
Webhook URLs are unique and secret - don't share them
If a webhook URL is compromised, delete and recreate the instant trigger
Use HTTPS only - AssistLoop rejects HTTP webhook URLs

Testing Your Integration

First, create a connection in Make to test authentication:

Add any AssistLoop module to your scenario
Click "Add" next to Connection
Enter Connection name: "AssistLoop Production"
Click "Continue" and authorize

The OAuth test will call GET /api/v1/oauth/me automatically.

Create a test scenario:

Add "AssistLoop: New Conversation" instant trigger
Click "Run once" to activate the webhook
Create a test conversation in your AssistLoop dashboard
Verify the data structure in Make's execution history

Add "AssistLoop: List Conversations" module
Select your workspace and agent
Set filters (status=1 for open conversations)
Run once and inspect the output

Get a conversation UUID from previous step
Add "AssistLoop: Find Messages" module
Set the conversation UUID
Add "AssistLoop: Reply to Conversation" module
Test sending a message

Add "AssistLoop: Close Conversation" module
Use a test conversation UUID
Run once and verify conversation is closed in dashboard

Troubleshooting

Common Issues

Getting Help

For Make integration support:

Make Help: https://www.make.com/en/help/tools
AssistLoop Documentation: https://assistloop.ai/docs
Support Email: [email protected]
Dashboard: https://app.assistloop.ai

When contacting support, please include:

Screenshot of your Make scenario
Execution history showing the error
Your organization UUID
Agent UUID (if applicable)
Full error message
Steps to reproduce

Module Reference Summary

Instant Triggers (Webhooks)

Module Name	Event Type	Description
New Conversation	new_conversation	Triggers when conversation starts
Human Handoff Requested	human_handoff_required	Triggers when AI needs human help

Actions

Module Name	Endpoint	Description
List Conversations	GET /conversations/list	Query conversations with filters
Find Messages	GET /messages/find	Get messages from a conversation
Reply to Conversation	POST /reply	Send a message to customer
Close Conversation	POST /close	Mark conversation as closed

Remote Procedures (RPCs)

RPC Name	Endpoint	Used In
Get Workspaces	GET /my/organizations	All modules (workspace selector)
Get Agents	GET /organizations/#{{id/agents	All modules (agent selector)
Get Conversations	GET /conversations/list	Find Messages, Reply, Close modules

This documentation provides everything you need to build powerful automations with AssistLoop and Make!

Make Integration

New Conversation

Human Handoff Required

General Endpoints

Close Conversation

Webhook Deliveries

Webhook Instant Trigger Not Firing

Conversation Queries Return Empty

Authentication Fails

Rate Limit Errors (429)

On this page