Groundng (QA For Cursor)

🚀 AI-driven Network Testing Tool

This project is an AI-driven network testing tool that automates the execution and management of network testing tasks through natural language instructions. It supports recording, executing, and discovering test steps, integrates with the MCP protocol, and can interact with configured AI coding assistants.

🚀 Quick Start

This AI-driven network testing tool allows you to automate network testing tasks using natural language instructions. Below is a quick guide to help you get started.

✨ Features

• Test Recording: Automatically perform browser operations based on natural language descriptions and save them in JSON format.
• Test Execution: Load recorded test cases and verify their results.
• Test Discovery: Analyze web page structures and generate suggestions for potential test steps.
• MCP Protocol Support: Can be directly invoked in the AI coding assistant through configuration.

📦 Installation

Prerequisites

• Python 3.10+
• LLM API access (Gemini 2.0 Flash is recommended)
• Installed MCP CLI tool (pip install mcp[cli])
• Playwright browser support (playwright install)

Installation Steps

1. Clone the repository:

git clone <repository-url>
cd <repository-name>

2. Create a virtual environment (recommended):

python -m venv venv
source venv/bin/activate # Linux/macOS
# venv\Scripts\activate # Windows

3. Install dependencies:

pip install -r requirements.txt

4. Install Playwright browser support:

playwright install --with-deps # Install browsers and system dependencies

📚 Documentation

Configuration Instructions

1. Rename the .env.example file in the project root directory to .env.
2. Add the LLM API key and other necessary information:

# .env
LLM_API_KEY="YOUR_LLM_API_KEY"

• Replace YOUR_LLM_API_KEY with your actual API key.

Usage Instructions

MCP Server Configuration

Add the following content to your MCP configuration:

{
  "mcpServers": {
    "Web-QA":{
      "command": "uv",
      "args": ["--directory","path/to/cloned_repo", "run", "mcp_server.py"]
    }
  }
}

Keep the server running to interact with the AI coding assistant.

Common Command Examples

• Record a test:

"Record a test: Visit https://practicetestautomation.com/practice-test-login/, enter 'student' into the username field, enter 'Password123' into the password field, click the submit button, and verify that the text 'Congratulations student' is visible."

• Execute a test:

"Run the regression test output/test_practice_test_login_20231105_103000.json"

• Discover test steps:

"Discover potential test steps starting from https://practicetestautomation.com/practice/."

• List recorded tests:

"List available web test records."

Output Instructions

• Recorded tests: Saved as JSON files in the output/ directory (see test_schema.md for the format).
• Execution results: Return a JSON object summarizing the running status (status, errors, evidence path). The full execution results are also saved to output/execution_result_....json.
• Discovery results: Return a JSON object containing the discovered URLs and suggested steps. The full discovery results are saved to output/discovery_results_....json.

🔧 Development and Contribution

Contributions are welcome! Please refer to CONTRIBUTING.md to learn how to get started, report issues, and submit pull requests.

📄 License

This project is licensed under APACHE-2.0.