OpenProject MCP Server
Enables AI agents to interact with OpenProject's APIv3 for autonomous project management, including task tracking, member administration, and project configuration. It supports comprehensive operations for managing work packages, projects, and reference data like statuses and priorities.
README
OpenProject MCP Server
A Model Context Protocol (MCP) server that enables AI agents to interact with OpenProject's APIv3 for autonomous project management.
Overview
This MCP server provides a comprehensive set of tools that allow an AI agent to:
- List, create, update, and delete projects
- Manage work packages (tasks, bugs, features, etc.)
- Handle project members and versions
- Query statuses, types, priorities, users, and roles
- Search across work packages
Features
Projects
- List all projects with pagination and filtering
- Get detailed project information
- Create new projects
- Update project details (name, description, status)
- Delete projects
- Manage project members (add/remove)
- Manage project versions
Work Packages
- List work packages with advanced filtering
- Get detailed work package information
- Create new work packages
- Update work packages (status, priority, assignee, dates, etc.)
- Delete work packages
- View activity history and comments
- Add comments to work packages
- Manage relations between work packages
Reference Data
- List all available statuses
- List all work package types
- List all priorities
- List all users
- List users available for assignment in a project
- List all roles
- List all groups
Search
- Full-text search across work packages
Prerequisites
- Node.js 18+
- An OpenProject instance with API access
Installation
# Clone or download this repository
cd openproject-mcp
# Install dependencies
npm install
# Build the TypeScript code
npm run build
Configuration
Set the following environment variables:
export OPENPROJECT_BASE_URL="https://your-openproject-instance.com"
export OPENPROJECT_API_KEY="your-api-key"
To get an API key:
- Log into OpenProject
- Go to My Account → API Access Keys
- Generate a new API key
Running
# Set environment variables first
export OPENPROJECT_BASE_URL="https://your-openproject-instance.com"
export OPENPROJECT_API_KEY="your-api-key"
# Run the server
node dist/index.js
Claude Desktop Integration
Add to your claude.json:
{
"mcpServers": {
"openproject": {
"command": "node",
"args": ["/path/to/openproject-mcp/dist/index.js"],
"env": {
"OPENPROJECT_BASE_URL": "https://your-openproject-instance.com",
"OPENPROJECT_API_KEY": "your-api-key"
}
}
}
}
Usage Example
Here's how an agent might use the tools:
- List projects to see what's available
- Get project types to understand what work package types exist
- Get statuses to know what transitions are possible
- List users to find assignable team members
- Create a work package with subject, type, priority
- Update the work package as work progresses (change status, update assignee)
- Add comments to document progress
- Search to find related work packages
Tool Reference
| Tool | Description |
|---|---|
op_list_projects |
List all projects |
op_get_project |
Get project details |
op_create_project |
Create a new project |
op_update_project |
Update project |
op_delete_project |
Delete a project |
op_list_project_members |
List project members |
op_add_project_member |
Add member to project |
op_remove_project_member |
Remove member from project |
op_list_project_versions |
List project versions |
op_create_version |
Create a version |
op_list_work_packages |
List work packages |
op_get_work_package |
Get work package details |
op_create_work_package |
Create a work package |
op_update_work_package |
Update work package |
op_delete_work_package |
Delete work package |
op_list_work_package_activities |
List activities |
op_add_work_package_comment |
Add comment |
op_list_work_package_relations |
List relations |
op_create_work_package_relation |
Create relation |
op_delete_work_package_relation |
Delete relation |
op_get_project_work_packages |
Get all WPs in project |
op_list_statuses |
List all statuses |
op_list_types |
List all types |
op_get_project_types |
List types in project |
op_list_priorities |
List all priorities |
op_list_users |
List all users |
op_get_user |
Get user details |
op_get_current_user |
Get current user |
op_list_project_assignees |
List assignable users |
op_list_roles |
List all roles |
op_list_groups |
List all groups |
op_search |
Search work packages |
Architecture
- client.ts: OpenProject API client using axios with Basic Auth
- index.ts: MCP server implementation with all tool handlers
License
MIT
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。