HP ALM MCP Server
Enables AI agents to interact with HP ALM / Quality Center for QA workflows including test case management, test execution, defect tracking, and requirements management.
README
HP ALM MCP Server
An MCP (Model Context Protocol) server that exposes HP ALM / Quality Center as tools for AI agents — GitHub Copilot, Claude Desktop, and any other MCP-compatible client.
It covers all day-to-day QA workflows: creating and managing test cases, pulling tests into test sets, recording execution results, filing defects, and managing requirements — all callable by an AI agent through natural language.
Table of Contents
- Features
- Available Tools
- Prerequisites
- Installation
- Configuration
- Running the Server
- Project Structure
- Security Notes
- Contributing
- License
Features
| Category | Count | What you can do |
|---|---|---|
| Session | 1 | Refresh / reconnect the ALM session |
| Test Plan — Folders | 2 | Create nested folder trees in Test Plan and Test Lab |
| Test Plan — Test Cases | 5 | List, get, find, create, bulk-create, update test cases |
| Test Plan — Version Control | 3 | Check out, check in, get VC status |
| Test Plan — Design Steps | 1 | Add / replace design steps |
| Test Lab — Test Sets | 2 | Find and create test sets |
| Test Lab — Test Instances | 3 | Add tests to sets, list instances, find by name |
| Test Execution | 5 | Create runs, update run/step status, full end-to-end execute |
| Defects | 4 | List, get, create, update defects |
| Requirements | 3 | List, get, create requirements |
| Attachments | 1 | Attach any file to any ALM entity |
| Search & Discovery | 2 | Generic HPQL search, list domains/projects |
| Total | 32 |
Available Tools
| Tool | Description |
|---|---|
alm_refresh_session |
Refresh or reconnect the ALM session |
alm_ensure_test_plan_folder |
Create nested Test Plan folder path |
alm_ensure_test_lab_folder |
Create nested Test Lab folder path |
alm_list_test_cases |
List test cases in a folder |
alm_get_test_case |
Get full details of a test case |
alm_find_test_by_name |
Find a test case ID by exact name |
alm_create_test_case |
Create a test case with optional design steps |
alm_update_test_case |
Update any field(s) on a test case |
alm_bulk_create_test_cases |
Create many test cases in one call |
alm_get_test_version_status |
Get VC status (Checked_In / Checked_Out) |
alm_checkout_test |
Check out a test case for editing |
alm_checkin_test |
Check in a test case after editing |
alm_add_design_steps |
Add / replace design steps on a test case |
alm_find_test_set |
Find a test set by name |
alm_create_test_set |
Create a test set in a Test Lab folder |
alm_add_test_to_set |
Pull a test from Test Plan into a test set |
alm_list_test_instances |
List all instances in a test set |
alm_find_test_instance |
Find a test instance by test case name |
alm_get_test_config |
Get test configuration ID for a test case |
alm_create_test_run |
Create a manual test run record |
alm_update_run_status |
Update the pass/fail status of a run |
alm_get_run_steps |
Get all run steps for a run |
alm_update_run_step |
Update status and actual result of a run step |
alm_execute_test |
Full end-to-end execution in one call |
alm_list_defects |
List defects with optional HPQL filter |
alm_get_defect |
Get full details of a defect |
alm_create_defect |
Create a new defect |
alm_update_defect |
Update any field(s) on a defect |
alm_list_requirements |
List requirements with optional HPQL filter |
alm_get_requirement |
Get full details of a requirement |
alm_create_requirement |
Create a new requirement |
alm_attach_to_entity |
Attach a local file to any ALM entity |
alm_search |
Generic HPQL search across any entity collection |
alm_list_domains_projects |
Discover all accessible domains and projects |
Prerequisites
- Python 3.11+
- An accessible HP ALM / Quality Center server (12.x, 15.x, 16.x tested)
- Network access from the machine running this server to your ALM instance
Installation
Option A — Install directly from GitHub (recommended for users)
pip install git+https://github.com/UditMahaldar/alm-mcp.git
Option B — Clone and install locally (recommended for contributors)
git clone https://github.com/UditMahaldar/alm-mcp.git
cd alm-mcp
pip install -e ".[dev]"
Configuration
All configuration is via environment variables or a .env file in the working directory.
# Copy the example and edit it
cp .env.example .env
| Variable | Required | Default | Description |
|---|---|---|---|
ALM_BASE_URL |
✅ | — | Base URL of your ALM server, e.g. https://alm.company.com/qcbin |
ALM_USERNAME |
✅ | — | ALM login username |
ALM_PASSWORD |
✅ | — | ALM login password |
ALM_DOMAIN |
✅ | — | ALM domain name |
ALM_PROJECT |
✅ | — | ALM project name |
ALM_REQUEST_DELAY |
❌ | 2.0 |
Seconds between API calls — increase if ALM throttles requests |
Security: Never commit your
.envfile. It is listed in.gitignoreby default.
Running the Server
Claude Desktop
Add this block to your Claude Desktop config file:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"hp-alm": {
"command": "python",
"args": ["-m", "alm_mcp.server"],
"env": {
"ALM_BASE_URL": "https://your-alm-server.example.com/qcbin",
"ALM_USERNAME": "your_username",
"ALM_PASSWORD": "your_password",
"ALM_DOMAIN": "YOUR_DOMAIN",
"ALM_PROJECT": "YOUR_PROJECT"
}
}
}
}
Restart Claude Desktop. You should see HP ALM MCP Server in the tools list.
VS Code / GitHub Copilot
Add this to your VS Code settings.json (or .vscode/mcp.json in the workspace):
{
"mcp": {
"servers": {
"hp-alm": {
"type": "stdio",
"command": "python",
"args": ["-m", "alm_mcp.server"],
"env": {
"ALM_BASE_URL": "https://your-alm-server.example.com/qcbin",
"ALM_USERNAME": "your_username",
"ALM_PASSWORD": "your_password",
"ALM_DOMAIN": "YOUR_DOMAIN",
"ALM_PROJECT": "YOUR_PROJECT"
}
}
}
}
}
Alternatively, if you installed via pip and want to use the entry-point script:
{
"mcp": {
"servers": {
"hp-alm": {
"type": "stdio",
"command": "alm-mcp",
"env": { "..." : "..." }
}
}
}
}
Standalone (stdio)
# With a .env file in the current directory
python -m alm_mcp.server
# Or with explicit env vars
ALM_BASE_URL=https://... ALM_USERNAME=user ALM_PASSWORD=pass \
ALM_DOMAIN=DEFAULT ALM_PROJECT=MyProject \
python -m alm_mcp.server
Project Structure
alm-mcp/
├── src/
│ └── alm_mcp/
│ ├── __init__.py # Package init
│ ├── config.py # pydantic-settings configuration
│ ├── alm_client.py # HP ALM REST API client
│ └── server.py # MCP tool definitions (FastMCP)
├── .env.example # Environment variable template
├── .gitignore
├── LICENSE
├── README.md
└── pyproject.toml
Security Notes
- SSL verification is disabled (
verify=False) because many enterprise ALM deployments use self-signed certificates. This is intentional and matches standard practice for on-premise ALM. - Credentials are never stored in code. They are loaded exclusively from environment variables or
.envfiles. - XML injection prevention: all user-supplied values are escaped with
html.escape()before being inserted into ALM XML payloads. - Path traversal prevention:
alm_attach_to_entityresolves and validatesfile_pathwithos.path.realpath()before opening the file. - No secrets in error messages: authentication error messages do not include the HTTP response body, preventing credential leakage.
Contributing
- Fork the repository and create a feature branch.
- Install dev dependencies:
pip install -e ".[dev]" - Make your changes.
- Open a pull request with a clear description.
License
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。