Testing Your API with Apidog

You have a working API. You understand how the code works. Now it's time to make sure everything actually works correctly - and Apidog's AI can help you test it thoroughly without writing test cases manually.

In this chapter, you'll learn how to use Apidog's AI features to automatically generate comprehensive test cases for your API, covering positive tests, negative tests, boundary cases, and security checks.

Before You Start: Enable AI Features

Apidog's AI features need to be enabled first. These features are free to use, but they require your own AI model API key (from Claude, OpenAI, or Gemini).

Enable AI features:

Go to Settings → AI Features

Toggle Enable AI Features to ON

Click + Add Provider to configure your AI model

Configure your AI provider:

Choose one of these providers and enter your API key:

Anthropic (Claude) - Recommended for best results

OpenAI (GPT-4)

Google AI Studio (Gemini)

Custom API (if using other providers)

Enter your API key and click Test to verify it works. If you don't have an API key yet:

Anthropic (Claude): Get one at console.anthropic.com

OpenAI (GPT-4): Get one at platform.openai.com

Google (Gemini): Get one at aistudio.google.com

Once configured, you'll see AI options throughout Apidog.

Setting Up Your Local Environment

Make sure your API is running locally:

You should see:

INFO:     Uvicorn running on http://127.0.0.1:8000
INFO:     Application startup complete.

In Apidog, create a Local environment:

Click Environments in the sidebar

+ New Environment

Name: "Local Development"

Base URL: http://localhost:8000

Save and switch to this environment

Now Apidog will send requests to your local API for testing.

Your First Manual Test

Before using AI, let's do one manual test so you understand what we're automating.

Test: Create a user

Open the POST /users endpoint in Apidog

Go to the Body tab

Enter test data:

{
  "email": "alice@example.com",
  "firstName": "Alice",
  "lastName": "Smith",
  "password": "SecurePass123",
  "phone": "+14155551234"
}

Click Send

Check the response:

Status code: 201 Created

Response includes id, email, firstName, createdAt

Response does NOT include password

That works. But what about edge cases? Invalid email? Missing fields? Duplicate email? You'd have to manually test all those scenarios.

This is where AI helps.

Generate Test Cases with AI

Instead of manually creating test cases for every possible scenario, let Apidog's AI generate them for you.

On the POST /users endpoint:

Click the Test Cases tab

Click Generate with AI button

A panel slides out on the right with options.

Select test case categories:

Check the types of tests you want:

✓ Positive tests - Valid inputs that should succeed

✓ Negative tests - Invalid inputs that should fail properly

✓ Boundary tests - Edge cases (empty strings, very long values, etc.)

✓ Security tests - SQL injection attempts, XSS, etc.

Configure credentials (if needed):

If your endpoint requires authentication, you can configure credentials here:

Add custom requirements:

In the text box at the bottom, you can add specific requirements to guide the AI:

Generate test cases that:
- Test duplicate email registration
- Test invalid email formats
- Test missing required fields
- Verify password is never returned in responses
- Test phone number format validation

Configure generation:

Set number of test cases: Up to 80 (try 20-30 for this endpoint)

Choose your AI model (if you configured multiple)

Click Generate and watch the AI work:

The AI analyzes your endpoint specification and generates comprehensive test cases.

Understanding Generated Test Cases

After generation completes, you'll see a list of test cases organized by category.

Positive Test Cases:

Create user with all valid fields

Valid email, firstName, lastName, password

Expects: 201 status, user object returned, no password in response

Create user without optional fields

Only required fields (email, firstName, lastName, password)

No phone number or preferences

Expects: 201 status, optional fields null or omitted

Negative Test Cases:

Create user with invalid email format

Email: "not-an-email"

Expects: 422 status, validation error message

Create user with missing required field

Missing firstName

Expects: 422 status, error indicating missing field

Create user with duplicate email

Email that already exists

Expects: 400 status, error about duplicate email

Boundary Test Cases:

Create user with very long email (255+ characters)

Tests email length limits

Expects: 422 status or truncation

Create user with empty string for firstName

Tests empty value handling

Expects: 422 status, validation error

Security Test Cases:

Create user with SQL injection in email

Email: admin'--@example.com

Expects: Properly escaped, no SQL injection

Create user with XSS attempt in firstName

firstName: <script>alert('xss')</script>

Expects: Properly sanitized or escaped

The AI generated all these based on your endpoint specification and common API testing patterns.

Running and Managing Test Cases

For each generated test case, you can review, run, and manage them:

Review a test case:

Click to expand and see request parameters, body, and expected response

You can rename it or adjust its category

Run it:

Click Run to execute the test against your local API

Apidog shows actual vs expected results

Response times are measured automatically

Accept good test cases:

If a test case is useful and passes correctly, click Accept

It gets saved to your Test Cases tab permanently

Discard irrelevant ones:

Some generated tests might not apply to your API

Click Discard to remove them

Bulk actions:

Select multiple test cases

Run, accept, or discard them all at once

This lets you quickly curate the best tests from the AI-generated batch.

What You've Learned

You now know how to:

Enable and configure AI features in Apidog

Generate comprehensive test cases automatically using AI

Select appropriate test categories (positive, negative, boundary, security)

Customize test generation with specific requirements

Run and manage AI-generated tests

Build test coverage without writing tests manually

The same process works for all your other endpoints (POST /user/login, GET /users/{id}, PUT /users/{id}, DELETE /users/{id}). Just open each endpoint, click "Generate with AI," select your test categories, and let the AI create comprehensive tests.

Most importantly, you can thoroughly test your API without spending hours writing test cases. The AI understands API testing patterns and generates relevant tests from your endpoint specifications.

What's Next

Your API is tested and you know it works correctly. Now it's time to prepare it for production deployment.

Continue with → Deployment

We'll deploy your API to a real server so it's accessible online, not just on your local machine.

Testing Your API with Apidog

Before You Start: Enable AI Features#

Setting Up Your Local Environment#

Your First Manual Test#

Generate Test Cases with AI#

Understanding Generated Test Cases#

Running and Managing Test Cases#

What You've Learned#

What's Next#