🤖 QA Copilot AI

An AI-powered QA Assistant that helps QA engineers analyze test failures, identify root causes, generate fixes, classify defects, detect flaky tests, and automate testing activities using MCP, TypeScript, and Large Language Models.

🚀 Project Overview

Modern QA teams spend significant time analyzing failed automation tests, debugging errors, writing bug reports, and identifying flaky tests.

QA Copilot AI is an AI-assisted testing platform designed to reduce debugging effort by combining:

• Model Context Protocol (MCP)

• Large Language Models (Groq LLM)

• Playwright automation data

• REST APIs

• TypeScript backend services

The system allows QA engineers to provide test reports and receive AI-generated insights such as:

• Root cause analysis

• Failure classification

• Suggested fixes

• Test improvement recommendations

• Automated QA artifacts

🎯 Problem Statement

Automation failures often require manual investigation:

❌ Reading long stack traces ❌ Finding root causes ❌ Identifying flaky tests ❌ Writing defect reports ❌ Creating repetitive test cases

QA Copilot AI solves this by acting as an intelligent testing assistant.

✨ Features

Related MCP server: BugForge

AI-Powered QA Analysis

✅ Failure Analysis ✅ Test Report Analysis ✅ Root Cause Identification ✅ Fix Suggestions ✅ Failure Classification ✅ Flaky Test Detection

Test Automation Assistance

✅ API Test Generator ✅ Test Case Generator ✅ Playwright Script Generator ✅ Bug Report Generator

Engineering Features

✅ MCP Server Implementation ✅ REST API using Express.js ✅ Modular TypeScript Architecture ✅ Environment Configuration ✅ Error Handling ✅ Logging Utilities

🏗 System Architecture

Workflow:

QA Engineer
     |
     |
Test Reports / Logs
     |
     v
QA Copilot API
     |
     v
MCP Server
     |
     |
-----------------------------
|      |        |            |
Failure Report  Fix     Flaky
Analyzer Analyzer Suggest Detector

     |
     v

Groq Large Language Model

     |
     v

AI Generated QA Insights

🛠 Tech Stack

📁 Project Structure

QA-Copilot-AI

src/
 ├── ai/
 │    └── promptTemplates.ts
 |
 ├── api/
 |
 ├── mcp-server/
 |
 ├── tools/
 |
 ├── services/
 |
 ├── config/
 |
 └── utils/

docs/
 ├── architecture.png
 └── architecture.md

screenshots/

package.json
README.md

⚙ Installation

Clone repository:

git clone https://github.com/Sham-coder16/qa-copilot-ai.git

Navigate:

cd qa-copilot-ai

Install dependencies:

npm install

🔑 Environment Setup

Create .env

GROQ_API_KEY=your_api_key
PORT=3000

▶ Running Application

Start MCP Server

npm run mcp

Expected:

MCP SERVER STARTED
Tools Registered Successfully

Start API Server

npm run api

Expected:

API Server running on port 3000

Run Test Client

npx tsx src/test-client.ts

🔧 MCP Tools

Available tools:

📡 REST API Example

Endpoint:

POST /analyze

Request:

{
 "failure": "Login test failed due to timeout"
}

Response:

{
 "category":"Automation Failure",
 "rootCause":"Element loading issue",
 "suggestion":"Increase explicit wait"
}

📸 Screenshots

📸 Project Screenshots

MCP Server Running

API Server Running

AI Failure Analysis

Bug Report Generation

Test Case Generation

🔄 Future Enhancements

• Docker Support

• GitHub Actions CI/CD

• Jenkins Pipeline Integration

• Database Storage

• Web Dashboard

• Multiple LLM Provider Support

• Playwright HTML Report Integration

💡 Skills Demonstrated

• QA Automation

• Playwright Testing

• TypeScript Development

• Node.js Backend

• REST API Development

• AI Integration

• Prompt Engineering

• MCP Architecture

• LLM Testing Concepts

• GitHub Project Management

👩‍💻 Author

Shamli Kadukar

GitHub:

https://github.com/Sham-coder16

⭐ If this project helped you understand AI-powered QA automation, consider starring the repository.