🧠 From Building Habits to Breaking Testing Barriers – My API Testing Journey with Keploy

Building APIs is fun — but testing them? That used to feel like a chore.
As part of the Keploy GitHub Learning Program, I recently built a full-stack Habit Tracker API using Flask and MongoDB — but the real shift came when I got to test it using AI.

📌 The Project: Habit Tracker API

The idea was simple — track personal habits for different users with full CRUD operations. I built:

A custom REST API in Flask

A MongoDB backend

A basic HTML + JavaScript frontend for testing

5 core endpoints: Add, Get, Update, and Delete habits

Sounds complete, right? Almost. But what about testing?
😫 The Problem: Manual API Testing

Initially, I tried writing unit tests manually — and it worked, but:

It was time-consuming

I struggled to cover all use cases

Every time I changed my API, I had to update my tests again

That’s when Keploy entered the scene.
🤖 The Game Changer: Keploy's AI Testing

With Keploy CLI, I was able to:

✅ Automatically record API calls
✅ Replay them as test cases
✅ Achieve real test coverage without manually scripting each test

I used curl commands and the frontend UI to simulate real use, while Keploy silently captured everything — and then turned those interactions into executable test cases.
Literally zero code for test writing!

🛠️ CI/CD Integration with GitHub Actions

Testing alone isn’t enough — it needs to be automated.

So I went ahead and:

Created an OpenAPI schema (using Flasgger)

Integrated Keploy test execution in GitHub Actions

Ensured my pipeline runs tests on every push and pull request

This step made my API production-ready in the truest sense.
🌐 Chrome Extension Bonus

To take it one step further, I used Keploy’s Chrome Extension to test APIs on real-world websites.
It felt amazing to inspect network traffic and capture meaningful test data without dev tools or Postman.

💡 What I Learned

Testing doesn’t have to be painful.

AI + DevTools = game-changing productivity

CI/CD is not complete without reliable test automation

Real coverage > theoretical coverage

🧩 Tech Stack Used

Backend: Flask (Python), MongoDB

Frontend: HTML + JS

Testing: Keploy CLI, Keploy Chrome Extension

CI/CD: GitHub Actions

🔗 See it in action:

👉 GitHub Repo Link – Habit Tracker API

Big shoutout to @keploy for building such a developer-friendly tool and helping students like me explore test automation with real-world projects.

Want to test APIs like a pro?
Try Keploy — you might just fall in love with testing 💙