Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/arainey2022/myaskai-docs/llms.txt

Use this file to discover all available pages before exploring further.

The User Data API lets your AI agent give personalised answers by fetching live data about the person it is talking to — their subscription tier, recent orders, account status, or any other information your backend holds. Rather than calling your systems directly, My AskAI calls a webhook URL you register. Your endpoint receives the user’s identifier, looks up the relevant data, and returns it as a JSON response. The agent then injects that context into its reasoning before composing a reply. This is especially powerful for e-commerce and SaaS support scenarios: instead of giving a generic “check your account page” reply, the agent can say “Your last order (#12345) was delivered on 1 September and is currently showing as Delivered.”
If you run a Shopify store, you can use the no-code Shopify connector instead of building a custom endpoint. Go to Knowledge → Content → Connect live user data to get started.

​
How It Works

1

You register a webhook URL

In the My AskAI dashboard go to Knowledge → Content, scroll to Connect live user data, and click + New Connection. Enter your endpoint URL, the Authorization header name, and the corresponding secret key.
2

A user starts a conversation

When a user with a validated email address contacts your AI agent, My AskAI detects their identifier (e.g. their email address, extracted from a Zendesk ticket or an authenticated Intercom conversation).
3

My AskAI calls your endpoint

My AskAI makes a POST request to your registered URL, sending the user’s identifier in the request body. Your endpoint must respond within 10 seconds.
4

Your endpoint returns user data

You return a JSON payload containing any user context that may be relevant for answering support questions — name, plan, order history, entitlements, and so on.
5

The agent uses the data

My AskAI injects the returned data into the agent’s context window. The agent can then reference it when composing its reply.

​
Configuring the Webhook

  1. Log in to your My AskAI dashboard.
  2. Go to Knowledge → Content.
  3. Scroll to Connect live user data and click + New Connection.
  4. Fill in the following fields:
    • API endpoint URL — the full URL My AskAI should call (e.g. https://api.yourcompany.com/myaskai/user-data)
    • Authorization Header — the header name your endpoint expects (e.g. Authorization)
    • Authorization Header Key — the secret value My AskAI will send in that header
  5. Enter a test user email address and click Test User Data API to verify your endpoint returns a valid response.
  6. Once satisfied, click the toggle to Activate live user data.

​
Incoming Request from My AskAI

My AskAI sends the following POST request to your endpoint for every new support conversation or ticket where the user’s identifier is available.

​
Headers My AskAI Sends

HeaderValue
Content-Typeapplication/json
AuthorizationThe secret key you configured in the dashboard

​
Request Body

​
identifier
string
required
The user’s email address, extracted from the authenticated support conversation or ticket. This is the lookup key your endpoint should use to retrieve user data.

​
Example Request Body

{
  "identifier": "alex@example.com"
}

​
Expected Response from Your Endpoint

Your endpoint must return an HTTP 200 with a JSON body containing the following fields.
​
user_info
string
required
A string containing all relevant user information. Format it clearly so the agent can parse and reference individual data points — plain text with newlines or a structured format both work well. Keep it focused: include only information that might be needed to answer a support question.
If user_info exceeds 1,500 characters, each response is charged as a tool call at $0.02 per message.
​
identifier
string
required
Echo back the same identifier value that was sent in the request.

​
Example Response

{
  "user_info": "Name: Alex Johnson\nEmail: alex@example.com\nSubscription: Pro (Monthly)\nSubscription Start Date: January 15, 2024\nNext Renewal Date: January 15, 2025\n\nRecent Orders:\n1. Order #12345\n   - Date: September 1, 2024\n   - Items: Running Shoes (x1)\n   - Status: Delivered\n\n2. Order #12346\n   - Date: September 10, 2024\n   - Items: Fitness Tracker (x1)\n   - Status: In Transit\n\n3. Order #12347\n   - Date: September 25, 2024\n   - Items: Wireless Earbuds (x1)\n   - Status: Processing",
  "identifier": "alex@example.com"
}

​
Authentication of Incoming Requests

My AskAI authenticates itself to your endpoint using the shared secret you configure in the dashboard. Verify this secret on every request to ensure you only process calls from My AskAI. 
// Example Node.js Express middleware
app.post('/myaskai/user-data', (req, res) => {
  const authHeader = req.headers['authorization'];
  if (authHeader !== process.env.MYASKAI_WEBHOOK_SECRET) {
    return res.status(401).json({ error: 'Unauthorized' });
  }
  // ... look up user data and respond
});

​
Performance Requirements

RequirementValue
Maximum response time10 seconds
Response formatapplication/json
Maximum user_info length (free)1,500 characters

​
Error Handling

Your endpoint should return appropriate HTTP status codes for error conditions. If your endpoint is unavailable or returns an error, My AskAI will continue the conversation without user context rather than failing the entire conversation.
Status codeMeaning
200 OKUser data returned successfully
400 Bad RequestRequest body missing the identifier field
401 UnauthorizedAuthorization header missing or incorrect
404 Not FoundNo user found for the provided identifier
500 Internal Server ErrorUnexpected error in your backend
If your endpoint times out or returns a 5xx error, the agent continues without user context. This means it gives a generic rather than personalised answer, but the conversation is not interrupted. Monitor your endpoint’s reliability to avoid degraded personalisation.

Build docs developers (and LLMs) love

Get started for freeTalk to us