Making a call before processing a message
A premessage webhook makes a call to an external service or application every time a customer submits input. The external service can process the message before it is processed by your assistant.
Add a premessage webhook to your assistant if you want the webhook to be triggered before each incoming message is processed by your assistant.
If you are using a custom channel, the premessage webhook works with the v2 /message
API only (stateless and stateful). For more information, see the API reference.
All built-in channel integrations use this API.
You can use a premessage webhook to do the following types of things:
- Translate the customer's input to the language that is used by your assistant.
- Check for and remove any personally identifiable information, such as an email address or social security number that a customer might submit.
You can use this webhook in coordination with the postmessage webhook. For example, the postmessage webhook can do things like translate the response back into the customer's language or add back information that was removed for privacy reasons. For more information, see Making a call after processing a message.
For environments where private endpoints are in use, keep in mind that a webhook sends traffic over the internet.
Defining the webhook
You can define one webhook URL to use for preprocessing every incoming message.
The programmatic call to the external service must meet these requirements:
- The call must be a POST HTTP request.
- The request body must be a JSON object (
Content-Type: application/json
). - The call must return in 30 seconds or less.
If your external service supports GET requests only, or if you need to specify URL parameters dynamically at run time, consider creating an intermediate service that accepts a POST request with a JSON payload that contains any runtime values. The intermediate service can then make a request to the target service, passing these values as URL parameters, and then return the response to the dialog.
Do not set up and test your webhook in a production environment where the assistant is deployed and is interacting with customers.
To add the webhook details, complete the following steps.
-
In your assistant, go to Environments and open the environment where you want to configure the webhook.
-
Click the icon to open the environment settings.
-
On the Environment settings page, click Pre-message webhook.
-
Or, if you're using the classic experience, open the Assistants page.
-
For the assistant you want to configure, click the icon, and then choose Settings.
-
Click Webhooks, then click Pre-message webhook.
-
-
Set the Pre-message webhook switch to Enabled.
-
Decide whether to return an error if the webhook call fails.
When enabled, everything stops until the preprocessing step is completed successfully.
-
If you have a critical preprocessing step that must be taken before you want to allow the message to be processed by the assistant, enable this setting.
Test the process that you are calling regularly so to check if it's down, and can change this setting to prevent all of your message calls from failing.
-
When this setting is disabled, the assistant ignores any errors that it encounters and continues to process the incoming message without taking the preprocessing step. If the preprocessing step is helpful but not critical, consider keeping this setting disabled.
-
-
In the URL field, add the URL for the external application to which you want to send HTTP POST request callouts.
For example, you might write a Cloud Functions web action that checks if a message is in a language other than English, and send it to the Language Translator service to convert it to English. Specify the URL for your web action, as in this example:
https://us-south.functions.cloud.ibm.com/api/v1/web/my_org_dev/default/translateToEnglish.json
You must specify a URL that uses the SSL protocol, so specify a URL that begins with
https
.You cannot use a webhook to call a Cloud Functions action that uses token-based Identity and Access Management (IAM) authentication. However, you can make a call to a Cloud Functions web action or a secured web action.
-
In the Secret field, add a private key to pass with the request that can be used to authenticate with the external service.
The key must be specified as a text string, such as
purple unicorn
. The maximum length is 1,024 characters. You cannot specify a context variable.It is the responsibility of the external service to check for and verify the secret. If the external service does not require a token, specify any string that you want. You cannot leave this field empty.
If you want to see the secret as you enter it, click the Show password icon before you start typing. After you save the secret, the string is replaced by asterisks and can't be viewed again.
For more information about how this field is used, see Webhook security.
-
In the Timeout field, specify the length of time (in seconds) you want the assistant to wait for a response from the webhook before it returns an error. The timeout duration cannot be shorter than 1 second or longer than 30 seconds.
-
In the Headers section, add any headers that you want to pass to the service one at a time by clicking Add header.
For example, if the external application that you call returns a response, it might be able to send a response in multiple different formats. The webhook requires that the response is formatted in JSON. The following table illustrates how to add a header that indicates that you want the resulting value to be returned in JSON format.
Header example Header name Header value Content-Type
application/json
The service automatically sends an
Authorization
header with a JWT; you do not need to add one. If you want to handle authorization yourself, add your own authorization header and it is used instead.After you save the header value, the string is replaced by asterisks and can't be viewed again.
Your webhook details are saved automatically.
Testing the webhook
Do extensive testing of your webhook before you enable it for an assistant that is being used in a production environment.
The webhook is triggered when a message is sent to your assistant to be processed.
If you enable the setting that returns an error when the webhook call fails, the processing of the assistant is halted entirely if the webhook encounters any issues. Test the process that you are calling regularly so to check if it's down, and can change this setting to prevent all of your message calls from failing.
If you call an Cloud Functions web action, you can use the logging capability in Cloud Functions to help you troubleshoot your code. You can download the command line interface, and then enable logging with the activation polling command.
Troubleshooting the webhook
The following error codes can help you track down the cause of issues you might encounter. If you have a web chat integration, for example, you know that your webhook has an issue if every test message you submit returns a message such as There is an error with the message you just sent, but feel free to ask me something else
.
If this message is displayed, use a REST API tool, such as cURL, to send a test /message
API request, so you can see the error code and the full message that is returned.
Error code and message | Description |
---|---|
422 Webhook responded with invalid JSON body | The webhook's HTTP response body could not be parsed as JSON. |
422 Error validating webhook response | The webhook's HTTP response body was not a valid /message body. |
422 Webhook responded with [500] status code |
There's a problem with the external service that you called. The code failed or the external server refused the request. |
500 Processor Exception : [connections to all backends failing] |
An error occurred in the webhook microservice. It could not connect to backend services. |
Webhook security
To authenticate the webhook request, verify the JSON Web Token (JWT) that is sent with the request. The webhook microservice automatically generates a JWT and sends it in the Authorization
header with each webhook call. It is your
responsibility to add code to the external service that verifies the JWT.
For example, if you specify purple unicorn
in the Secret field, you might add code such as:
const jwt = require('jsonwebtoken');
...
const token = request.headers.authentication; // grab the "Authentication" header
try {
const decoded = jwt.verify(token, 'purple unicorn');
} catch(err) {
// error thrown if token is invalid
}
Request body
It is useful to know the format of the request body of the premessage webhook so that your external code can process it.
The payload contains the request body of the /message
(stateful or stateless) v2 API request. The event name message_received
indicates that the request is generated by the premessage webhook. For more information about
the message request body, see the API reference.
{
"payload" : { Copy of request body sent to /message }
"event": {
"name": "message_received"
}
}
Response body
The service that receives the POST request from the webhook must return a JSON object (Accept: application/json
).
The response body must have the following structure:
{
"payload": {
...
}
}
The payload
object in the response should contain the payload
object that was received in the request body. Your code can modify property values in the message payload it received (for example, to update property values,
or to add or remove context variables); but the message payload that is returned to the service must conform to the schema for a request to the message
method. For more information, see the API reference.
Example 1
This example shows you how to check the language of the input text, and append the language information to the input text string.
In the premessage webhook configuration page, the following values are specified:
- URL:
https://us-south.functions.appdomain.cloud/api/v1/web/e97d2516-5ce4-4fd9-9d05-acc3dd8ennn/default/check_language
- Secret: none
- Header name: Content-Type
- Header value: application/json
The premessage webhook calls an IBM Cloud Functions web action name check_language
.
The node.js code in the check_language
web action looks as follows.
let rp = require("request-promise");
function main(params) {
console.log(JSON.stringify(params))
if (params.payload.input.text !== '') {
// Send a request to the Watson Language Translator service to check the language of the input text.
const options = { method: 'POST',
url: 'https://api.us-south.language-translator.watson.cloud.ibm.com/instances/572b37be-09f4-4704-b693-3bc63869nnnn/v3/identify?version=2018-05-01',
auth: {
'username': 'apikey',
'password': 'nnn'
},
headers: {
"Content-Type":"text/plain"
},
body: [
params.payload.input.text
],
json: true,
};
return rp(options)
.then(res => {
params.payload.context.skills["actions skill"].user_defined["language"] = res.languages[0].language;
console.log(JSON.stringify(params))
//Append "in" plus "the language code" to the input text, surrounded by parentheses.
const response = {
body : {
payload : {
input : {
text : params.payload.input.text + ' ' + '(in ' + res.languages[0].language + ')'
},
},
},
};
return response;
})
}
return {
body : params
}
};
To test the webhook, click Preview. Submit the text Buenos días
. The assistant probably can't understand the input, and returns the response from your Anything else node. However, if you go to the Analyze
page of your assistant and open Conversations, you can see what was submitted. Check the most recent user conversation. The log shows that the user input is Buenos días (in es)
. The es
in parentheses
represents the language code for Spanish, so the webhook worked and recognized that the submitted text was a Spanish phrase.
Example 2
This example shows you how to check the language of the incoming message, and if it's not English, translate it into English before you submit it to the assistant.
Define a sequence of web actions in IBM Cloud Functions. The first action in the sequence checks the language of the incoming text. The second action in the sequence translates the text from its original language into English.
In the premessage webhook configuration page, the following values are specified:
- URL:
https://us-south.functions.appdomain.cloud/api/v1/web/e97d2516-5ce4-4fd9-9d05-acc3dd8ennn/default/translation_sequence
- Secret: none
- Header name: Content-Type
- Header value: application/json
The node.js code for the first web action in your sequence looks as follows:
let rp = require("request-promise");
function main(params) {
console.log(JSON.stringify(params))
if (params.payload.input.text !== '') {
const options = { method: 'POST',
url: 'https://api.us-south.language-translator.watson.cloud.ibm.com/instances/572b37be-09f4-4704-b693-3bc63869nnnn/v3/identify?version=2018-05-01',
auth: {
'username': 'apikey',
'password': 'nnn'
},
headers: {
"Content-Type":"text/plain"
},
body: [
params.payload.input.text
],
json: true,
};
return rp(options)
.then(res => {
//Set the language property of the incoming message to the language that was identified by Watson Language Translator.
params.payload.context.skills["actions skill"].user_defined["language"] = res.languages[0].language;
console.log(JSON.stringify(params))
return params;
})
}
else {
params.payload.context.skills["actions skill"].user_defined["language"] = 'none'
return params
}
};
The second web action in the sequence sends the text to the Watson Language Translator service to translate the input text from the language that was identified in the previous web action into English. The translated string is then sent to your assistant instead of the original text.
The node.js code for the second action in your sequence looks as follows:
let rp = require("request-promise");
function main(params) {
console.log(JSON.stringify(params))
//If the the incoming message is not null and is not English, translate it.
if ((params.payload.context.skills["actions skill"].user_defined.language !== 'en') && (params.payload.context.skills["actions skill"].user_defined.language !== 'none')) {
const options = { method: 'POST',
url: 'https://api.us-south.language-translator.watson.cloud.ibm.com/instances/572b37be-09f4-4704-b693-3bc63869nnnn/v3/translate?version=2018-05-01',
auth: {
'username': 'apikey',
'password': 'nnn'
},
headers: {
"Content-Type":"application/json"
},
//The body includes the parameters that are required by the Language Translator service, the text to translate and the target language to translate it into.
body: {
text: [
params.payload.input.text
],
target: 'en'
},
json: true
};
return rp(options)
.then(res => {
params.payload.context.skills["actions skill"].user_defined["original_input"] = params.payload.input.text;
const response = {
body : {
payload : {
"context" : params.payload.context,
"input" : {
"text" : res.translations[0].translation,
"options" : {
"export" : true
}
},
},
},
};
return response
})
}
return {
body : params
}
};
When you test the webhook in the preview panel, you can submit Buenos días
, and the assistant responds as if you said Good morning
in English. In fact, when you check the Analyze page of your assistant and open Conversations,
the log shows that the user input was Good morning
.
You can add a postmessage webhook to translate the message's response back into the customer's language before it is displayed. For more information, see Example 2.
Removing the webhook
If you decide you do not want to preprocess customer input with a webhook, complete the following steps:
-
In your assistant, go to Environments and open the environment where you want to remove the webhook.
-
Click the icon to open the environment settings.
-
On the Environment settings page, click Pre-message webhook.
-
Or, if you're using the classic experience, open the Assistants page.
-
For the assistant you want to configure, click the icon, and then choose Settings.
-
Click Webhooks, then click Pre-message webhook.
-
-
Do one of the following things:
-
To change the webhook that you want to call, click the Delete webhook button to delete the currently specified URL and secret. You can then add a URL and other details.
-
To stop calling a webhook to process every incoming message, click the switch to disable the webhook altogether.
-