kao273183

kao273183 / mk-qa-master

Public

AI 測試大師 — MCP server driving pytest / Jest / Cypress / Go / Maestro. Analyze, generate, run, advise. Web + Mobile (iOS/Android/BlueStacks).

mcp.chenjundigital.com maestro mcp mcp-server mobile-testing model-context-protocol
13
1
89% credibility
Found May 17, 2026 at 16 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

MK QA Master is an AI-powered testing assistant that transforms how developers and QA engineers approach automated testing. By connecting to your favorite AI assistant, it can analyze your web pages or mobile screens to discover what needs testing, automatically generate runnable test code, execute tests across multiple frameworks (web and mobile), and provide intelligent suggestions for improvement. The tool creates a continuous feedback loop where every test run teaches the system what to fix next, making your testing suite smarter over time. It's designed for teams who want to move faster without sacrificing quality, handling everything from simple smoke tests to complex multi-platform regression suites.

How It Works

1
💡 You want to test your app but don't know where to start

You have a website or mobile app and need automated tests, but writing them from scratch feels overwhelming.

2
🔌 You connect your AI assistant to your testing project

With a simple setup, your AI assistant becomes your testing partner, ready to run and create tests for you.

3
🔍 Your AI explores your app and maps out what to test

Give it a URL and it automatically finds forms, buttons, and API connections—discovering everything that needs testing.

4
✨ Your AI writes ready-to-run tests for you

Based on what it discovered, it generates actual test code with real selectors—not empty placeholders.

5
Your app works on different platforms
🌐
Web testing

Test websites using Playwright with screenshots and step-by-step recordings

📱
Mobile testing

Test iOS and Android apps using Maestro with flows that work on any device

6
▶️ You run your tests and watch them pass

Your AI executes the tests and captures everything—passes, failures, screenshots, and videos.

🎯 You get a beautiful report and a plan to improve

See your test results in a gorgeous dashboard, and receive smart suggestions on what to fix or add next.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 16 to 13 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is mk-qa-master?

MK QA Master is an MCP server written in Python that brings AI-driven test automation to your existing workflows. It connects to AI coding assistants like Claude Desktop or Cursor via the Model Context Protocol and drives testing across pytest, Jest, Cypress, Go test, and Maestro for mobile. The core workflow: paste a URL, and it probes the DOM, maps out forms and buttons, then generates runnable test code with real selectors—no empty stubs. After running tests, it archives results, surfaces flaky tests, and produces a prioritized action plan telling you exactly what to fix next.

Why is it gaining traction?

The URL-to-test generation is the hook. Most test generators produce skeleton code you still have to manually complete. This one reads the actual page structure and pre-fills the selectors. The self-improvement loop is a second differentiator: it tracks test history across runs, flags patterns in flaky tests, and measures whether generated tests actually get adopted. The multi-runner support means web teams and mobile teams can use the same MCP tool surface without learning separate integrations.

Who should use this?

Frontend teams using Playwright who want AI help generating regression tests from live pages. QA engineers managing test suites across pytest and Jest who need automated coaching on what to prioritize next. Mobile teams testing iOS/Android via Maestro who want the same AI-assisted workflow as their web counterparts. It is less suited for teams already deep in a proprietary test framework or those needing native CI dashboard integrations.

Verdict

The credibility score of 0.9% and 13 stars reflect a young, solo-built project—but the documentation is thorough, the architecture is clean, and the URL-to-runnable-test feature works. Try it via `uvx mk-qa-master` before committing, especially if you are evaluating AI-assisted QA tooling. The foundations are strong; the community backing just needs time to build.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.