External Identities self-service sign-up validation using a Python Azure Function and API connector

This sample demonstrates how to use API connectors to customize sign-up for Azure AD guest user self-service sign-up and Azure AD B2C sign-up user flows.

In particular, the sample demonstrates how to:

  1. Limit external user sign-ups to only a particular email domain. In this example, it's a fictitious fabrikam.com and fabricam.com.
  2. Validate a user-provided value ('Job Title') against a validation rule.

The API is implemented using an Azure Function HTTP trigger in Python.


File/folder Description
SignUpValidation/init.py Sample source code for HTTP trigger.
.gitignore Define what to ignore at commit time.
CHANGELOG.md List of changes to the sample.
CONTRIBUTING.md Guidelines for contributing to the sample.
README.md This README file.
LICENSE The license for the sample.

Key concepts

API connectors provide you with a way to modify and extend sign-up flows by leveraging web APIs. API connectors are available in both guest user self-service sign up and Azure AD B2C sign-up user flows.

This examples uses an API connector to limit sign-ups to only specific email domains, fabrikam.com and fabricam.com. This is easily modifiable in __init__.py and can be extended limit sign ups to any particular email domain or set of email domains. Further, the API connector in this sample is used to perform input validation on 'Job Title' by ensuring a user provides a value of at least 4 characters.

This sample uses an Azure Function as the web API endpoint but you can alternatively edit the .__init__.py file in your preferred IDE and deploy that code in any web service. If so, environment variables used for authentication may work differently.


Before you get started, make sure you have the following requirements in place:

Setting up your Azure Function

Steps to run locally

  1. Clone the repository
git clone https://github.com/Azure-Samples/active-directory-python-external-identities-api-connector-azure-function-validate
  1. Navigate to the Azure extension in Visual Studio code on the left navigation bar. You should see a 'Local Project' folder representing your local Azure Function.
  2. Press F5 (or use the Debug > Start Debugging menu command) to launch the debugger and attach to the Azure Functions host. (This command automatically uses the single debug configuration that Azure Functions created.)
  3. The Azure Function extension will automatically generate a few files for local development, install dependencies, and install the Function Core tools if not already present. These tools help with the debugging experience.
  4. Output from the Functions Core tools appears in the VS Code Terminal panel. Once the host has started, Alt+click the local URL shown in the output to open the browser and run the function. You can also see the url of the Local Function by right clicking on the function on the Azure Functions explorer.
  5. To redeploy the local instance during testing, just repeat these steps.

Add authentication

Authentication is stored in environment variables, so they're not stored as part of the repository and should never be stored in checked in code. Read more about the local.settings.json file.

  1. Create a local.settings.json file
  3. You final local.settings.json should look like following one:
  "IsEncrypted": false,
  "Values": {
    "AzureWebJobsStorage": "",

Specify a Username and Password. This will be what your Azure Function uses to authenticate incoming requests from Azure AD.

Deploy the application

  1. Follow steps of this guide to deploy your Azure Function to the cloud. Copy the endpoint web URL of your Azure Function.
  2. Once deployed, you'll see a 'Upload settings' option. Select this. It will upload your environment variables onto the Application settings of function app.

To learn more about Visual Studio Code development for Azure Functions, see this.

Configure and enable the API connector

Follow the steps outlined in "Add an API connector" for guest user self-service sign-up or for Azure AD B2C to create an API connector and enable it your user flow. The end result is shown below.

API connector configuration

Your API connector configuration should look like the following:

API connector configuration

  • Endpoint URL is the Function URL you copied earlier.
  • Username and Password are the Username and Passwords you defined as environment variables earlier.

Enable the API connector

In the API connector settings for your user flow, you can select the API connector to be invoked at either step: API connector selected

  • After signing in with an identity provider - if enabled for this step, the API connector will only allow users with an email ending in @fabrikam.com. Note that for Azure AD B2C, this does not apply to local accounts.
  • Before creating the user - if enabled for this step, the API connector will only allow users with an email ending in @fabrikam.com and check whether 'Job Title' is of at least length 4. Note that Job Title has to be selected in User attributes for the user flow.

Customizing the Azure Function

This sample provides a quick way to get started using API connectors. By modifying the source code and leveraging all the capabilities of a web API you're used to, you'll be able to accomplish many more complex scenarios including integration with other web APIs and services.


This project welcomes contributions and suggestions. Most contributions require you to agree to a Contributor License Agreement (CLA) declaring that you have the right to, and actually do, grant us the rights to use your contribution. For details, visit https://cla.opensource.microsoft.com.

When you submit a pull request, a CLA bot will automatically determine whether you need to provide a CLA and decorate the PR appropriately (e.g., status check, comment). Simply follow the instructions provided by the bot. You will only need to do this once across all repos using our CLA.

This project has adopted the Microsoft Open Source Code of Conduct. For more information see the Code of Conduct FAQ or contact opencode@microsoft.com with any additional questions or comments.