FxKit MCP Server
Provides comprehensive documentation, code examples, and implementation assistance for the FxKit C# functional programming library. It enables AI models to search API references and access pre-configured prompts for functional patterns like Option, Result, and railway-oriented programming.
README
FxKit MCP Server
A Model Context Protocol (MCP) server that provides comprehensive documentation and assistance for the FxKit C# functional programming library.
Overview
FxKit MCP dynamically fetches and serves documentation from the official FxKit documentation site, providing AI models with:
- Complete documentation for all FxKit types (Option, Result, Validation, Unit)
- Compiler services documentation (source generators, analyzers)
- Code examples and API references
- Search capabilities across all documentation
- Pre-built prompts for common FxKit patterns
Features
Resources
The server exposes FxKit documentation sections as MCP resources:
introduction- FxKit overview and conceptsgetting-started- Installation and setup guideoption- Option type documentationresult- Result type documentationvalidation- Validation type documentationunit- Unit type documentationcompiler-overview- Compiler services overviewenum-match- EnumMatch generatorunion- Union type generatorlambda- Lambda generatortransformer- Transformer generatortesting- Testing utilities
Tools
Interactive tools for querying FxKit documentation:
search_documentation- Search across all documentationget_examples- Get code examples for specific featureslist_packages- List available NuGet packages
Prompts
Pre-configured prompts for common FxKit patterns:
convert-nullable-to-option- Convert nullable types to Optionrailway-oriented-programming- Implement railway-oriented patternsvalidation-multiple-errors- Use Validation for error accumulationgenerate-union-types- Create union types with source generators
Installation
From Source
- Clone this repository:
git clone https://github.com/taxfyle/fxkit-mcp.git
cd fxkit-mcp
- Install dependencies:
npm install
- Build the project:
npm run build
Global Installation
npm install -g @taxfyle/fxkit-mcp
Usage
With Claude Code
{
"mcpServers": {
"fxkit-docs": {
"command": "npx",
"args": [
"-y",
"@taxfyle/fxkit-mcp@latest"
],
"env": {}
}
}
}
With Claude Desktop
Add the following to your Claude Desktop configuration file:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"fxkit-docs": {
"command": "node",
"args": ["/path/to/fxkit-mcp/dist/index.js"]
}
}
}
Or if installed globally:
{
"mcpServers": {
"fxkit-docs": {
"command": "fxkit-mcp"
}
}
}
With MCP Inspector
Test the server using the MCP Inspector:
npx @modelcontextprotocol/inspector tsx src/index.ts
Programmatic Usage
import { FxKitMcpServer } from 'fxkit-mcp';
const server = new FxKitMcpServer();
await server.start();
Development
Running in Development Mode
npm run dev
Project Structure
fxkit-mcp/
├── src/
│ ├── index.ts # Main entry point
│ ├── server.ts # MCP server implementation
│ ├── types.ts # TypeScript type definitions
│ ├── fetcher/
│ │ └── docs-fetcher.ts # Documentation fetching logic
│ ├── resources/ # Resource handlers (if extended)
│ ├── tools/ # Tool implementations (if extended)
│ └── prompts/ # Prompt templates (if extended)
├── dist/ # Compiled JavaScript
├── package.json
├── tsconfig.json
└── README.md
Testing
The server includes a test mode that verifies all documentation endpoints are accessible:
npm test
Examples
Using with an AI Assistant
Once configured, you can ask your AI assistant questions like:
- "Show me the FxKit documentation for the Result type"
- "Search for examples of railway-oriented programming with FxKit"
- "How do I convert nullable types to Option in FxKit?"
- "Generate a union type for payment methods using FxKit"
The AI will use the MCP server to fetch relevant documentation and provide accurate, up-to-date information about FxKit.
Architecture
The server uses a dynamic fetching approach:
- No Static Content: Documentation is fetched on-demand from the live FxKit documentation site
- Caching: Responses are cached for 5 minutes to improve performance
- HTML Parsing: Uses Cheerio to extract content and code examples from documentation pages
- Type Safety: Full TypeScript implementation with strict typing
Contributing
Contributions are welcome! Please feel free to submit issues or pull requests.
Adding New Documentation Sections
To add new documentation sections:
- Update the paths array in
src/fetcher/docs-fetcher.ts - Add corresponding resource registration in
src/server.ts - Test that the new section is accessible
Extending Tools
New tools can be added in src/server.ts in the setupTools() method. Follow the existing pattern using registerTool().
License
MIT
Acknowledgments
- FxKit - The functional programming library for C#
- Model Context Protocol - The protocol specification
- effect-mcp - Inspiration for MCP server structure
Support
For issues related to:
- FxKit library: Visit the FxKit GitHub repository
- This MCP server: Open an issue in this repository
- MCP protocol: Check the MCP documentation# fxkit-mcp
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。