Skip to content

checkly/checkly-sample-pom

Repository files navigation

Checkly CLI PageObject example

This example project shows how you can use the Checkly CLI in a monitoring as code (MaC) workflow using the PageObject Model (POM). We are using https://danube-web.shop as a monitoring target.

  1. Write API Checks and Playwright-powered Browser Checks!
  2. Test -> Deploy: now you have your app monitored around the clock. All from your code base.

Project Structure

This project shows one possible way to build POM based checks that can be executed on-demand and run on a schedule. One flow, checkout, is covered right now, with more coming soon.

.
|-- __checks__
|   |-- key-user-flows
|   |   |-- pages
|   |      |-- cart.ts
|   |      |-- ...
|   |-- api.check.ts
|   `-- checkout.spec.ts
|-- checkly.config.ts
|-- playwright.config.ts
|-- README.md
`-- package.json
  • Running npx checkly test will look for .check.ts files and .spec.ts in __checks__ directories and execute them in a dry run.

  • Running npx check deploy will deploy your checks to Checkly, attach alert channels, and run them on a 10m schedule in the region us-east-1 and eu-west-1

  • As an additional option, running npx playwright test will run your .spec.ts in __checks__ directories locally using Playwright with the settings specified in playwright.config.ts

CLI Commands

Run the core CLI commands with npx checkly <command>

Command Action
npx checkly test Dry run all the checks in your project
npx checkly deploy Deploy your checks to the Checkly cloud
npx checkly login Log in to your Checkly account
npx checkly --help Show help for each command.

Check the docs for the full CLI reference.

tip: you can add autocomplete to your shell using npx checkly autocomplete

About

No description, website, or topics provided.

Resources

Code of conduct

Security policy

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published