What should I use for API smoke tests in a pipeline?

For API smoke tests in a pipeline, use a small, fast, deterministic set of checks that verify the API is basically alive and the most critical flows work.

Good choices

1. Postman/Newman

• Easy to author and maintain
• Good for quick validation of a few endpoints
• Runs well in CI/CD via Newman
• Best when your team already uses Postman collections

2. pytest + requests or JUnit + RestAssured

• Better if you want tests in the same language as your app
• More flexible and maintainable for code-heavy teams
• Good for assertions, setup/teardown, and integration with existing test suites

3. cURL + shell scripts

• Simple and lightweight
• Good for the most basic health checks
• Less maintainable once the smoke suite grows

4. Karate

• Strong option if you want readable API tests with less code
• Good for smoke + functional API testing
• Nice balance between simplicity and power

What smoke tests should cover

Keep them minimal:

• GET /health or equivalent
• Authentication works
• One critical read endpoint
• One critical write endpoint
• Basic dependency check if relevant

What to avoid

• Full regression suites
• Expensive end-to-end flows
• Tests that depend on unstable data
• Long-running or flaky assertions

Best practice

Use one small smoke suite that:

• runs on every deploy
• finishes in under a few minutes
• fails fast
• uses dedicated test data or isolated environments

Practical recommendation

• If you want fastest setup: Postman + Newman
• If you want best maintainability in code: pytest/requests or RestAssured
• If you want a good middle ground: Karate

If you tell me your stack (language, CI tool, and whether you already use Postman), I can recommend the best option more specifically.