Product Thumbnail

KushoAI

Open-source AI tester that lives in your CLI

Open Source
Developer Tools
Artificial Intelligence
GitHub

An open-source AI Agent that lives in your terminal. Generate UI tests with one command, run them instantly or plug into your CI pipeline. No boilerplate, no brittle scripts — just reliable testing built by developers, for developers.

Top comment

Hi Hunters, Makers, and Users 👋,

I’m Abhishek, Co-founder of @KushoAI.

KushoAI is already used by tens of thousands of developers and QA engineers across global companies. We’re an AI platform that removes busywork from software development so teams can spend more time shipping.

A lot of you asked for an agent that's scriptable, git-friendly, and usable without leaving the terminal. So we built it.

Meet the KushoAI CLI Agent, an open-source AI tester that lives in your terminal and writes UI tests for you.

What it’s optimized for:

Generate UI tests for any web interface with a single command

🧠 Thinks up realistic scenarios you’d spend hours designing, covering flows, edge cases, and validations tailored to your UI

🧮 Broad coverage fast, exploring permutations and variations to maximize confidence without brittle hand-written scripts

🔄 Run inside CI/CD for continuous coverage with every commit

🧰 Terminal-native workflow, so you can generate, review, commit, and run tests all in your CLI

🌍 Open-source and extensible, fork it, customize it, and contribute on GitHub

We started with agents for API, UI, security, and end-to-end backend workflow testing. The CLI is the next step for developers who want a scriptable agent that plugs directly into repos and pipelines.

👉 Star & contribute on GitHub
👉 Explore the full platform

Would love your thoughts, feedback, and PRs. 🚀

Comment highlights

Curious how stable this is in real CI/CD runs—does it handle flaky tests better than Playwright alone? IMO that’s the real bottleneck in most pipelines.

Congrats on the launch! 🙌

KushoAI looks like a real game-changer for QA teams. I’d love to know—how does it handle complex workflows where API and UI tests need to be combined?