drupalorg-mcp
MCP server for drupal.org's public REST API. Enables querying projects, issues, comments, and user information on drupal.org.
README
drupalorg-mcp
MCP server for drupal.org's public api-d7 REST API. Read-only, no authentication required.
Quick start (Claude Code)
claude mcp add drupalorg -s user -e DRUPALORG_USERNAME=your-username -- npx -y drupalorg-mcp
-e DRUPALORG_USERNAME=... is optional — set it to enable author: "me" in tool calls instead of passing your username every time. Verify with claude mcp list.
Tools
- resolve_project — resolve a project machine name (e.g.
commerce) to its numeric project ID. - resolve_user — resolve a drupal.org username to its numeric uid.
- list_issues — list issues, optionally filtered by project, status, priority, category, and/or author. Omit the project to search across all of drupal.org (e.g. "issues I created"). Paginated (50/page).
- get_issue — full details for a single issue by node ID, including related merge requests and credit metadata.
- get_issue_comments — all comments for an issue by node ID.
- list_my_projects — list projects (modules/themes/distributions/core/recipes) owned by a given user. Useful before listing "open issues on my project".
- get_project_maintainers — list a project's maintainers and their permissions, by machine name.
- find_maintained_projects — reverse maintainer lookup, scoped to one owner: given a user and an owner (e.g.
centarro), returns which of the owner's projects that user maintains. drupal.org has no global user→projects maintainer endpoint, so this scansmaintainers.jsonfor every project authored byowner— not usable unscoped across all of drupal.org.
Tools that take an author/username/owner argument accept a username, a numeric uid, or the literal "me" (requires DRUPALORG_USERNAME env var, see below).
Requirements
- Node.js 18+ (uses global
fetch)
Usage
Runs via npx — no install/clone needed, always fetches the latest published version.
Add to other MCP clients
E.g. Claude Desktop's config:
{
"mcpServers": {
"drupalorg": {
"command": "npx",
"args": ["-y", "drupalorg-mcp"],
"env": { "DRUPALORG_USERNAME": "your-username" }
}
}
}
DRUPALORG_USERNAME is optional — set it to enable author: "me" in tool calls instead of passing your username every time.
Running from a local checkout
For development, or if you'd rather not fetch from npm on every launch:
npm install
node index.js
See CONTRIBUTING.md for the development workflow.
Notes
- The
UAheader inindex.jsidentifies this tool to drupal.org's API — update it if you fork/republish under a different repo. - All status/priority/category values are translated from drupal.org's internal numeric IDs to readable labels.
Contributing
See CONTRIBUTING.md for development setup, code structure, and publishing steps.
Built with Claude Code — code, docs, CI workflow, and npm packaging were written by Claude in collaboration with the repo owner.
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。