Automation Framework

Automate web, APIs, mobile and performance with one framework.

k0lmena is an open source test automation framework built on Playwright. Write scenarios in Gherkin, implement them in TypeScript and get shareable HTML reports.

Get it on GitHub Implement it in my company

open source · ISC
TypeScript
Playwright
Cucumber / BDD
Appium
k6 · Artillery

bash

$ git clone https://github.com/underc0delabs/k0lmena.git
$ npm install
$ npx playwright install
$ cp .example.env .env
$ K0LMENA_AUTO_HEALING=on npm run test

  chromium · firefox · webkit
  ✓ 42 passed  0 failed
  HEALED: #btn-login → role=button[name="Ingresar"]
$ npm run report

4test types

3browsers

1syntax for everything

$0in licenses

what's inside

Four kinds of tests in the same repository

Same structure, same syntax and same reports, whatever the engine underneath.

Type	Engine	What it covers
Front / E2E	Playwright + Cucumber	UI on Chromium, Firefox and WebKit.
API	Axios + Cucumber	HTTP endpoints with response validation.
Mobile	WebdriverIO + Appium	Mobile apps locally, on emulators or on BrowserStack.
Performance	Artillery + k6	Load, stress, soak and spike, over HTTP or the browser.

features

What it saves you every day

Locator auto-healing

Learns alternative selectors when an action succeeds and uses them when the original breaks. Every heal is logged in the report.

HTML reports per type

Front, API, mobile and performance each produce their own report, including Artillery Cloud.

Evidence on failure

Screenshots, browser console logs and Playwright traces attached automatically.

Parallel execution

Run front and API suites with N workers or one per CPU using a single variable.

Helper tools

Locator generator, broken-link tester, Playwright codegen recorder and debug mode.

CI/CD ready

GitHub Actions workflow on the official Playwright container, with reports as artifacts.

how a test is written

Scenarios the whole team can read

Scenarios are written in Gherkin and steps use reusable helpers with auto-healing built in.

src/front-test/features/login.feature

Feature: Login
  @Smoke
  Scenario: Sign in with a valid user
    Given I am on the login page
    When I fill the user with "[email protected]"
    And I click on "Ingresar"
    Then I see the dashboard

src/front-test/steps/login.tests.ts

When("I fill the user with {string}",
  async (user) => {
    await getByPlaceholderAndFillIt(
      page, "User", user);
  });

When("I click on {string}",
  async (name) => {
    await getElementByRoleAndClickIt(
      page, "button", name);
  });

See how to add a new test

video

See it in action

A walkthrough of installation, project structure and test runs on the QARMY channel.

Watch on YouTube Read the docs

k0lmena Automation Framework video tutorial

for companies

We implement it with your team

We install the framework in your repository, automate your critical flows, wire it into your CI/CD and train your people. All tailored to you.

Setup

Project structure, per-environment config, Cucumber profiles and conventions for your team.

Critical flow automation

Web, API and mobile smoke and regression suites, plus performance tests with your business thresholds.

CI/CD

A pipeline with parallel runs, reports as artifacts and evidence on every failure.

Training

Hands-on training so your team writes and maintains its own tests.

Request a proposal

faq

Frequently asked questions

What is k0lmena Automation Framework?

It's an open source test automation framework built on Playwright and TypeScript that brings UI, API, mobile and performance tests into one repository, with Gherkin (BDD) scenarios and HTML reports.

Is it free?

Yes. It's open source under the ISC license and available on GitHub.

What do I need to use it?

Node.js (LTS recommended) and optionally Visual Studio Code. For mobile, Appium and an emulator or device (or a BrowserStack account). For k6, a bundled script installs the binary.

What is auto-healing?

It recovers broken locators. In learn mode the framework stores alternative selectors for each element; in heal mode, if the original selector fails, it tries the candidates until one works and logs it in the report.

Can it run in CI/CD?

Yes. It ships with a GitHub Actions workflow that runs front, API and performance tests in the official Playwright container and publishes the reports as artifacts.

How does it relate to K0lmenIA?

K0lmenIA ships with k0lmena built in: its mapper agents walk through your app and write the .feature files, steps and locators for you. The suite then runs with npm, without AI and with zero tokens.

let's talk

Want to automate your tests with k0lmena?

Tell us what you want to automate and we'll help you get it running in your team.

Talk to us on WhatsApp Join the community