magnolia-docs-mcp
Enables AI assistants to search and retrieve information from the official Magnolia CMS documentation through a set of MCP tools.
README
Magnolia CMS Documentation MCP Server (unofficial)
An MCP (Model Context Protocol) server that provides tools to search and retrieve information from the official Magnolia CMS documentation.
Overview
This MCP server allows AI assistants and language models to search the Magnolia CMS documentation directly, making it easier to get accurate and up-to-date information about Magnolia CMS features, configuration, and best practices.
Features
- Documentation Search: Search through the official Magnolia CMS documentation
- Structured Results: Returns formatted search results with titles, URLs, and snippets
- MCP Compatible: Works with any MCP-compatible client or AI assistant
Installation
- Clone or download this repository
- Install dependencies:
npm install
Usage
As a Standalone MCP Server
This server is designed to run as an MCP server using the stdio transport. You can test it directly:
node src/main.js
With MCP-Compatible Clients
To use this server with MCP-compatible clients (like Claude Desktop, Continue, or other AI assistants), you need to configure the client to connect to this server.
Example Configuration for Claude Desktop
Add the following to your MCP configuration file (usually ~/.config/claude-desktop/mcp.json or similar):
{
"mcpServers": {
"magnolia-docs": {
"command": "node",
"args": ["/path/to/your/project/src/main.js"],
"cwd": "/path/to/your/project"
}
}
}
Example Configuration for Other MCP Clients
The server can be configured as a stdio-based MCP server in any compatible client. Refer to your client's documentation for specific configuration instructions.
Available Tools
search_magnolia_docs
Searches the official Magnolia CMS documentation for the given query.
Parameters:
query(string, required): The search term or phrase
Returns:
- An array of search results containing:
title: The title of the documentation pageurl: The URL to the documentation pagesnippet: A brief excerpt from the page
Example Usage:
{
"name": "search_magnolia_docs",
"arguments": {
"query": "content types configuration"
}
}
Project Structure
magnolia-docs-mcp/
├── src/main.js # Main MCP server implementation
├── package.json # Project dependencies and metadata
└── README.md # This file
Dependencies
@modelcontextprotocol/sdk: MCP SDK for building MCP serverscheerio: HTML parsing for extracting search resultsnode-fetch: HTTP client for making requests to the documentation site
Development
Modifying Search Behavior
The search functionality is implemented in the search_magnolia_docs tool handler in main.js. You can modify:
- Search URL construction
- HTML parsing selectors (currently uses
.search-result,.title,.snippet) - Number of results returned (currently limited to 5)
- Result formatting
Adding New Tools
To add new tools:
- Add the tool definition to the
ListToolsRequestSchemahandler - Add the tool implementation to the
CallToolRequestSchemahandler - Update this README with the new tool documentation
Testing
You can test the server by running it directly and sending MCP requests via stdio:
node src/main.js
Then send JSON-RPC requests according to the MCP specification.
Error Handling
The server includes error handling for:
- Network failures when accessing the documentation site
- HTML parsing errors
- Invalid tool requests
Errors are returned as MCP error responses with descriptive messages.
License
MIT
Author
Daniel Naranjo
Contributing
- Fork the repository
- Create a feature branch
- Make your changes
- Test your changes
- Submit a pull request
Troubleshooting
Common Issues
- "Module not found" errors: Make sure you've run
npm installto install dependencies - Connection issues: Verify that the Magnolia CMS documentation site is accessible
- No search results: The HTML selectors might need updating if the documentation site structure changes
Debugging
Enable verbose logging by modifying the error handlers in main.js to include more detailed error information.
Roadmap
- [ ] Add support for different documentation sections
- [ ] Implement caching for better performance
- [ ] Add configuration options for search parameters
- [ ] Support for multiple documentation versions
- [ ] Enhanced result filtering and ranking
Related Resources
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。
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。
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。
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。
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。
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。
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。
mcp-server-qdrant
这个仓库展示了如何为向量搜索引擎 Qdrant 创建一个 MCP (Managed Control Plane) 服务器的示例。