WHartTest
WHartTest 是基于 Django REST Framework 与现代大模型技术打造的 AI 驱动测试自动化平台。平台聚合自然语言理解、知识库检索与嵌入搜索能力,结合 LangChain 与 MCP(Model Context Protocol) 工具调用,实现从需求到可执行测试用例的自动化生成与管理,帮助测试团队提升效率与覆盖率。
claude mcp add --transport stdio mgdaaslab-wharttest python manage.py runserver 0.0.0.0:8913 \ --env DJANGO_SETTINGS_MODULE="WHartTest.settings"
How to use
WHartTest is an AI-driven automated testing platform built on Django REST Framework. It leverages LangChain and MCP (Model Context Protocol) to orchestrate AI-assisted generation of test cases, manage projects and requirements, review tasks, and organize a knowledge base for enhanced testing guidance. The system exposes an API-backed interface and an admin UI, enabling teams to generate, organize, and refine test cases with AI assistance. The MCP integration allows you to invoke language-model tools and external services within the test-generation workflow to produce high-quality, coverage-oriented test scenarios tailored to your project needs.
To use the MCP capabilities, start the Django server and interact with the platform’s endpoints to initiate AI-driven test case generation, manage knowledge-base entries, and review or modify generated test cases. The MCP integration makes it possible to call external tools or services as part of the AI planning and execution, enabling more accurate data extraction, embedding-based retrieval, and customized tool usage within your test automation pipeline.
How to install
Prerequisites:
- Python 3.8+ and pip
- Git
- Optional: Docker and docker-compose for containerized deployment
Installation steps:
-
Clone the repository git clone https://github.com/MGdaasLab/WHartTest.git cd WHartTest
-
Create and activate a virtual environment (recommended) python -m venv venv
Windows
venv\Scripts\activate
macOS/Linux
source venv/bin/activate
-
Install dependencies pip install -r requirements.txt
-
Prepare configuration cp .env.example .env # if you have environment-specific settings
Ensure your OpenAI/Azure/Ollama keys and other settings are configured in .env or WHartTest/settings as needed
-
Run the server locally python manage.py migrate python manage.py collectstatic --noinput python manage.py runserver 0.0.0.0:8913
-
(Optional) Docker deployment
- Use the provided Docker deployment guidance in the documentation to deploy with docker-compose and expose port 8913.
- Ensure the .env file is configured inside the container for API keys and model endpoints.
Additional notes
Tips and caveats:
- The server exposes an admin interface at /admin; default admin credentials are admin/admin123456 (please change in production).
- For production, remove default API keys and configure secure authentication and TLS.
- If using MCP tool calls, ensure external tools or models are accessible from the server environment and that required API keys are set in the environment.
- The port 8913 is used for the Django app; adjust firewall rules accordingly if hosting behind a reverse proxy.
- Refer to the project docs for detailed MCP tool usage, knowledge base integration, and deployment guides.
Related MCP Servers
mcp-vegalite
MCP server from isaacwasserman/mcp-vegalite-server
github-chat
A Model Context Protocol (MCP) for analyzing and querying GitHub repositories using the GitHub Chat API.
nautex
MCP server for guiding Coding Agents via end-to-end requirements to implementation plan pipeline
pagerduty
PagerDuty's official local MCP (Model Context Protocol) server which provides tools to interact with your PagerDuty account directly from your MCP-enabled client.
futu-stock
mcp server for futuniuniu stock
mcp -boilerplate
Boilerplate using one of the 'better' ways to build MCP Servers. Written using FastMCP