Definition:
Karate DSL is an open-source test automation framework that combines API testing, UI testing, and performance testing into a single, readable, programming-language-free domain-specific language. Instead of writing test code in Python, Java, or JavaScript, you write tests in Gherkin-style plain text (Given/When/Then syntax) that describes exactly what the test should do — and Karate runs it without you writing a single line of programming code.
The revolutionary idea behind Karate: API tests should NOT require programming knowledge. Product owners, QA analysts, and developers should all be able to read, write, and review the same test files.
═══════════════════════════════════════════════════════════════
WHAT KARATE DSL REPLACES
═══════════════════════════════════════════════════════════════
Traditional API Test (Python + Requests library):
┌─────────────────────────────────────────────────────────┐
│ │
│ import requests │
│ import json │
│ │
│ def test_create_user(): │
│ payload = { │
│ "name": "Alice", │
│ "email": "alice@test.com", │
│ "role": "admin" │
│ } │
│ response = requests.post( │
│ "https://api.example.com/users", │
│ json=payload, │
│ headers={"Authorization": "Bearer xxx"} │
│ ) │
│ assert response.status_code == 201 │
│ data = response.json() │
│ assert data["name"] == "Alice" │
│ assert data["role"] == "admin" │
│ assert data["id"] is not None │
│ │
│ → Requires: Python knowledge, HTTP knowledge, │
│ JSON parsing, assertion syntax │
│ → Only readable by: developers │
│ │
└─────────────────────────────────────────────────────────┘
Same Test in Karate DSL:
┌─────────────────────────────────────────────────────────┐
│ │
│ Feature: User Management │
│ │
│ Scenario: Create a new admin user │
│ Given url 'https://api.example.com' │
│ And path '/users' │
│ And header Authorization = 'Bearer xxx' │
│ And request { name: 'Alice', email: 'alice@test.com',│
│ role: 'admin' } │
│ When method post │
│ Then status 201 │
│ And match $.name == 'Alice' │
│ And match $.role == 'admin' │
│ And match $.id != '#null' │
│ │
│ → Requires: No programming knowledge │
│ → Readable by: Devs, QAs, PMs, product owners │
│ → Each line is self-explanatory English │
│ │
└─────────────────────────────────────────────────────────┘Example:
A visual walkthrough of a complete test scenario in Karate DSL — showing an API test, a data-driven test, and a mock server — all without writing any programming code.
SCENARIO: An e-commerce API. You need to test:
- Creating a product (POST)
- Looking up a product by ID (GET)
- Testing with MULTIPLE product types (data-driven)
- Mocking the payment service for isolated testing
═══════════════════════════════════════════════════════════════
TEST 1: API CRUD — Create and Verify a Product
═══════════════════════════════════════════════════════════════
Karate DSL test file (products.feature):
┌─────────────────────────────────────────────────────────┐
│ │
│ Feature: Product API │
│ │
│ Background: │
│ * url 'https://api.shop.example.com' │
│ * header Authorization = 'Bearer test-token-123' │
│ │
│ Scenario: Create a new product and verify it exists │
│ │
│ # Step 1: Create the product │
│ Given path '/products' │
│ And request { │
│ 'name': 'Wireless Keyboard', │
│ 'price': 49.99, │
│ 'category': 'electronics', │
│ 'inStock': true │
│ } │
│ When method post │
│ Then status 201 │
│ │
│ # Save the returned ID for later use │
│ And def productId = $.id │
│ │
│ # Step 2: Look up the product by ID │
│ Given path '/products', productId │
│ When method get │
│ Then status 200 │
│ │
│ # Verify the response │
│ And match $.name == 'Wireless Keyboard' │
│ And match $.price == 49.99 │
│ And match $.inStock == true │
│ And match $.category == 'electronics' │
│ And match $ contains { createdBy: '#notnull' } │
│ │
└─────────────────────────────────────────────────────────┘
═══════════════════════════════════════════════════════════════
TEST 2: DATA-DRIVEN — Same Test, Multiple Product Types
═══════════════════════════════════════════════════════════════
Instead of writing 8 separate test scenarios, use a
Scenario Outline with Examples:
┌─────────────────────────────────────────────────────────┐
│ │
│ Scenario Outline: Create products of different types │
│ │
│ Given path '/products' │
│ And request { │
│ 'name': '<name>', │
│ 'price': <price>, │
│ 'category': '<category>' │
│ } │
│ When method post │
│ Then status 201 │
│ And match $.category == '<category>' │
│ And match $.price == <price> │
│ And match $.id != '#null' │
│ │
│ Examples: │
│ | name | price | category | │
│ | Wireless Keyboard| 49.99 | electronics | │
│ | Running Shoes | 89.99 | sports | │
│ | Programming Book | 39.99 | books | │
│ | Desk Lamp | 24.99 | home | │
│ | Coffee Maker | 59.99 | kitchen | │
│ | Yoga Mat | 19.99 | sports | │
│ | Water Bottle | 14.99 | sports | │
│ | USB-C Cable | 9.99 | electronics | │
│ │
│ → One Scenario Outline runs 8 tests with different │
│ data. Add more rows to test more cases. │
│ → JSON/CSV files can also be used as data sources. │
│ │
└─────────────────────────────────────────────────────────┘
═══════════════════════════════════════════════════════════════
TEST 3: MOCKING — Simulate Payment Service Without Real $
═══════════════════════════════════════════════════════════════
Karate has a built-in mock server. No need for WireMock
or separate mock services.
┌─────────────────────────────────────────────────────────┐
│ │
│ Mock file (payment-mock.feature): │
│ │
│ Feature: Payment Service Mock │
│ │
│ Scenario: Simulate successful payment │
│ Given pathMatches('/payments/charge') │
│ And method 'post' │
│ And request.amount < 1000 │
│ Then status 200 │
│ And response { │
│ 'status': 'approved', │
│ 'transactionId': '#uuid' │
│ } │
│ │
│ Scenario: Simulate declined payment │
│ Given pathMatches('/payments/charge') │
│ And method 'post' │
│ And request.amount >= 1000 │
│ Then status 402 │
│ And response { │
│ 'status': 'declined', │
│ 'reason': 'amount_exceeds_limit' │
│ } │
│ │
│ Start the mock server: │
│ java -jar karate.jar -m payment-mock.feature -p 8089 │
│ │
│ → Now your real app can point to localhost:8089 │
│ and get realistic payment responses — FOR FREE. │
│ → No credit cards charged. No payment gateway needed. │
│ → Test both approved AND declined paths easily. │
│ │
└─────────────────────────────────────────────────────────┘
═══════════════════════════════════════════════════════════════
THE FULL TEST PIPELINE (How Karate DSL runs)
═══════════════════════════════════════════════════════════════
┌─────────────────────────────────────────────────────────┐
│ │
│ WRITE: Plain text .feature files │
│ (Given/When/Then syntax — no code) │
│ │
│ ↓ │
│ │
│ PARSE: Karate engine reads .feature files │
│ → Builds an internal test plan │
│ → Handles HTTP, auth, JSON automatically │
│ │
│ ↓ │
│ │
│ EXECUTE: │
│ ┌──────────────────────────────────────────────┐ │
│ │ API Tests → Sends HTTP requests │ │
│ │ Validates JSON/XML responses │ │
│ │ │ │
│ │ UI Tests → Drives browser via WebDriver │ │
│ │ (if using Karate UI module) │ │
│ │ │ │
│ │ Perf Tests → Converts to Gatling simulation │ │
│ │ (if using Karate Gatling) │ │
│ │ │ │
│ │ Mocks → Starts embedded HTTP mock │ │
│ │ server │ │
│ └──────────────────────────────────────────────┘ │
│ │
│ ↓ │
│ │
│ REPORT: HTML reports with request/response logs │
│ Timelines, assertions, coverage analytics │
│ CI integration (JUnit XML, Cucumber reports) │
│ │
└─────────────────────────────────────────────────────────┘