Skip to main content

PagerDuty

In this example, we will show you how you can get started working with the PagerDuty API to produce a working use case that lists the most recent reported incidents.

Before you start

You'll need a PagerDuty API key. For testing purposes you can get one directly from their API documentation.

Install & Authorize Superface

The Superface CLI provides all the tooling needed to author the Comlinks for your integration.

Install with Homebrew
brew install superfaceai/cli/superface

If you don't have a Superface account already, you can sign up here. You can use your account to authenticate the CLI.

superface login

Prepare the documentation

PagerDuty's API is available as an Open API Specification, you can use this as the documentation source.

superface prepare https://github.com/PagerDuty/api-schema/blob/main/reference/REST/openapiv3.json "pagerduty"

Once the documentation has been indexed, you will need to edit the pagerduty.provider.json that has been created to ensure that the correct API base URL and security policy are used. The exact provider definition you need to use is below:

pagerduty.provider.json
{
"name": "pagerduty",
"defaultService": "default",
"parameters": [],
"services": [
{
"baseUrl": "https://api.pagerduty.com",
"id": "default"
}
],
"securitySchemes": [
{
"type": "apiKey",
"in": "header",
"name": "Authorization",
"id": "pagerdutyToken"
}
]
}

Define your use case

Next, you can create a profile for the use case you want to achieve, which in this case is to get a list of companies in the CRM.

superface new pagerduty "get incidents"

A file called incident-management.get-incidents.profile will be created in the superface folder as a result of this. This is a Comlink file that decribes the input and output expectations that the API expects to achieve this use case.

To turn the Comlink profile into runnable code in either Node.js or Python, use the following command and include the name of the provider and a use case profile you want to work with.

superface map pagerduty incident-management/get-incidents

Add your API key

If you don't already have your PagerDuty API key in your environment, you can add it to your .env file as PAGERDUTY_API_KEY="Token token=<your-pagerduty-api-key>".

Set up your inputs

In order to list the incidents correctly, you will need to modify the boilerplate code from the PagerDuty documentation so it contains a real request object. From the superface folder, open incident-management.get-incidents.pagerduty.mjs (or .py if you created Python files) and modify the object in the usecase.perform() function.

incident-management.get-incidents.pagerduty.mjs
{
time_zone : 'UTC',
since : '2022-01-01T00:00:00Z',
until : '2022-01-31T23:59:59Z',
date_range : 'all',
sort_by : 'created_at',
sort_direction : 'asc',
offset: 0,
limit: 25,
},

Test it

To test it you can use the execute command.

superface execute pagerduty incident-management/get-incidents

If everything is successful you should see output similar to the response below:

RESULT: {
"incidents": [
{
"acknowledgements": [],
"alert_counts": {
"all": 0,
"resolved": 0,
"triggered": 0
}...