TestMu AI Test Manager MCP
Enables management of TestMu AI test projects, test cases, test runs, and integration with Jira, HyperExecute, and AI insights through natural language.
README
TestMu AI Test Manager MCP
An MCP (Model Context Protocol) server for TestMu AI Test Manager, HyperExecute, and AI Insights,
built with the official @modelcontextprotocol/sdk.
New to MCP, or setting this up for the first time? See GETTING_STARTED.md for a step-by-step walkthrough, including how to connect Claude Desktop, Claude Code, or another MCP client.
What this server provides
Tools are organized by domain under src/tools/:
- projects/, folders/ - TestMu AI Test Manager projects and test-case folders.
- testCases/ - create/read/update test cases and their execution history.
- testRuns/ - test runs and test-run folders, per-instance status/steps, bulk updates.
- jira/ - link/unlink Jira issues, execution history by Jira ID.
- environments/ - browser/OS/device environment lookup.
- users/ - organization user lookup (for assignees).
- attachments/ - file uploads for test steps/instances.
- hyperexecute/ - HyperExecute job/task/scenario/session execution detail.
- insights/ - AI-powered root cause analysis (RCA) and enriched test execution data.
Prerequisites
- Node.js 22+
- npm
- A TestMu AI account with Test Manager access
Installation
npm install
Configuration
Copy the example environment file and fill in your TestMu AI credentials:
cp .env.example .env
| Variable | Description | Default |
|---|---|---|
LT_USERNAME |
Your LambdaTest username | — |
LT_ACCESS_KEY |
Your LambdaTest access key | — |
LT_TM_BASE_URL |
Base URL for the LambdaTest Test Manager API | https://test-manager-api.lambdatest.com |
LT_ORG_ID |
Your LambdaTest account/org ID - optional, only needed by tm.link_jiraIssue |
— |
Build & Run
npm run build
npm start
Development
Run directly from TypeScript source without a build step:
npm run dev
The server communicates over stdio, which is how MCP clients (e.g. Claude Desktop, Claude Code) launch and talk to it.
Project Structure
src/
index.ts # Executable entry point: wires client + server + stdio transport
config.ts # Environment variable loading and validation (dotenv + zod)
server.ts # Constructs the McpServer instance and registers tools
client.ts # Reusable HTTP client (get/post/patch/delete/postForm) for calling the TestMu AI API
config/
endpoints.ts # Centralized registry of API endpoint paths - tools never hardcode a path
utils/
response.ts # Shared defensive-parsing helpers for reading API responses
tools/
index.ts # Central place where all tools are registered
serverInfo.ts # Orientation tool (tm.get_serverInfo)
projects/, folders/, testCases/, jira/, testRuns/, environments/, users/,
attachments/, hyperexecute/, insights/
# One domain per subfolder, one tool per file
Extending with New Tools
See CONTRIBUTING.md for the full set of conventions this project's tools follow (naming, file layout, input validation, response parsing, error handling, and rules for what belongs in a tool's own description vs. internal dev notes) - read it before adding a new tool, so the server keeps evolving consistently.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。