This repository contains a service that exposes an endpoint for anonymizing Personally Identifiable Information (PII). It uses the Microsoft Presidio project for the anonymization process. The application is built on Bun, a JavaScript runtime that helps to create performant applications.

The application exposes the /anonymize endpoint which accepts a POST request. This endpoint simplify the process by doing the analyze and the anonimize process at once. The request body should contain a text to be anonymized, a language shortcode, and optionally, a configuration for the anonymizers.

While this sample demo use English only, Presidio can be used to detect PII entities in multiple languages. Refer to the multi-language support for more information.

The request body should contain a text to be anonymized, and optionally, a score, an exlude array (words in that array will not be ananymized) and a configuration for the anonymizers. You can find more info regarding the available operators in Presidio documentation here: https://microsoft.github.io/presidio/anonymizer/#built-in-operators And a list of the supported entities is available here: https://microsoft.github.io/presidio/supported_entities/

Here is an example of a simple request: POST: http://127.0.0.1:3006/anonymize

{
	"text":"John Smith phone number is +33123456789"
}

The response will contain the anonymized text:

  {
	"text": "<PERSON> phone number is <PHONE_NUMBER>",
	"items": [
		{
			"start": 25,
			"end": 39,
			"entity_type": "PHONE_NUMBER",
			"text": "<PHONE_NUMBER>",
			"operator": "replace"
		},
		{
			"start": 0,
			"end": 8,
			"entity_type": "PERSON",
			"text": "<PERSON>",
			"operator": "replace"
		}
	]
}

And a more complex one with anonymizers configuration:

{
	"text":"John Smith phone number is +33123456789",
	"language":"en",
	"score":0.75,
	"exclude":["min", "hi", "earth", "max", "Voiceflow"],
	"anonymizers": {
		"PHONE_NUMBER": {
			"type": "mask",
			"masking_char": "*",
			"chars_to_mask": 8,
			"from_end": false
		},
		"EMAIL_ADDRESS": {
			"type": "mask",
			"masking_char": "*",
			"chars_to_mask": 8,
			"from_end": false
		},
		"IP_ADDRESS": {
			"type": "mask",
			"masking_char": "*"
		},
		"DATE_TIME": {
			"type": "keep"
		},
		"URL": {
			"type": "keep"
		},
		"LOCATION": {
			"type": "replace",
			"new_value": "LOCATION"
		},
		"DEFAULT": {
			"type": "replace",
			"new_value": "ANONYMIZED"
		}
	}
}

The response:

  {
	"text": "ANONYMIZED phone number is ******456789",
	"items": [
		{
			"start": 27,
			"end": 39,
			"entity_type": "PHONE_NUMBER",
			"text": "******456789",
			"operator": "mask"
		},
		{
			"start": 0,
			"end": 10,
			"entity_type": "PERSON",
			"text": "ANONYMIZED",
			"operator": "replace"
		}
	]
}

The application uses the following environment variables:

• NODE_ENV: The environment in which the application is running (default to "production").
• PORT: The port on which the application will listen.
• ANALYZE_ENDPOINT: The endpoint of the Presidio Analyzer service (default to http://presidio-analyzer:3000/analyze).
• ANONYMIZE_ENDPOINT: The endpoint of the Presidio Anonymizer service (default to http://presidio-anonymizer:3000/anonymize).
• SCORE: The score threshold for the Presidio Anonymizer service (default to 0.75).

These variables should be set in a .env file in the root directory of the project. You can use the .env.template as an example:

NODE_ENV = production
PORT = 3006
ANALYZE_ENDPOINT = http://presidio-analyzer:3000/analyze
ANONYMIZE_ENDPOINT = http://presidio-anonymizer:3000/anonymize
SCORE = 0.75

The application use Docker Compose to manage the presidio services, the app and its dependencies. The docker-compose.yml file in the root directory of the project defines the services needed to run the application.

The package.json file defines several scripts for running the application and managing Docker containers:

• dev: Runs only the application and in development mode, with file watching enabled.
• app: Runs only the application
• start: Starts all the containers and the application.
• stop: Stops all the containers and the application.

So to run everything, use: npm start

https://youtu.be/CKIoCGraulM

You can import the demo project on your Voiceflow dashboard using this link: https://creator.voiceflow.com/dashboard?import=652911763fb9db0007fca037

We can talk about this project on Discord https://discord.gg/9JRv5buT39