MCP 服务器

Content Manager MCP Server

Enables comprehensive content management through Markdown processing, HTML rendering, intelligent fuzzy search, and document analysis. Supports frontmatter parsing, tag-based filtering, table of contents generation, and directory statistics for efficient content organization and discovery.

README

Content Manager MCP Server

A comprehensive TypeScript-based Model Context Protocol (MCP) server for content management, including Markdown processing, document generation, and intelligent note searching.

🚀 Features

📝 Markdown Processing

Render Markdown to HTML with table of contents generation
Extract headings and create structured navigation
Parse frontmatter from Markdown files
HTML sanitization for safe content processing

🔍 Intelligent Search

Fuzzy search through notes using Fuse.js
Exact text matching with context highlighting
Tag-based filtering using frontmatter metadata
Date range queries for time-based content discovery

📊 Content Analysis

Directory statistics showing file counts, sizes, and types
File discovery with customizable extension filtering
Content preview with metadata extraction

🛠 Tech Stack

TypeScript - Type-safe development with latest ES2023 features
Zod - Runtime schema validation and type inference
tsup - Fast TypeScript bundler with ESM support
pnpm - Efficient package management
Fuse.js - Fuzzy search functionality
marked - Markdown parsing and rendering

📦 Installation

Prerequisites

Node.js >= 20.0.0
pnpm >= 9.0.0

Setup

# Clone and install dependencies
git clone <repository-url>
cd content-manager-mcp
pnpm install

# Build the project
pnpm build

# Run in development mode
pnpm dev

🔧 Usage

As MCP Server

The server runs via stdio and implements the MCP protocol:

node dist/index.js

Available Tools

`render_markdown`

Converts Markdown content to HTML with optional features:

{
  content: string,           // Markdown content to render
  generateToc?: boolean,     // Generate table of contents (default: true)
  sanitizeHtml?: boolean,    // Sanitize output HTML (default: true)
  enableCodeHighlight?: boolean // Enable syntax highlighting (default: true)
}

`search_notes`

Intelligent search through content files:

{
  query: string,             // Search query
  directory?: string,        // Directory to search (default: cwd)
  includeContent?: boolean,  // Include file content in results (default: true)
  maxResults?: number,       // Maximum results to return (default: 10)
  fuzzy?: boolean           // Use fuzzy search (default: true)
}

`search_by_tags`

Find files by frontmatter tags:

{
  tags: string[],           // Array of tags to search for
  directory?: string        // Directory to search (default: cwd)
}

`search_by_date_range`

Find files by modification date:

{
  startDate: string,        // ISO date string (YYYY-MM-DD)
  endDate: string,          // ISO date string (YYYY-MM-DD)
  directory?: string        // Directory to search (default: cwd)
}

`generate_table_of_contents`

Extract and format headings as a table of contents:

{
  content: string           // Markdown content to analyze
}

`extract_headings`

Get structured heading information:

{
  content: string           // Markdown content to analyze
}

`extract_frontmatter`

Parse YAML frontmatter from Markdown:

{
  content: string           // Markdown content with frontmatter
}

`get_directory_stats`

Analyze directory content statistics:

{
  directory?: string        // Directory to analyze (default: cwd)
}

`find_content_files`

Discover files with specific extensions:

{
  directory?: string,           // Directory to search (default: cwd)
  extensions?: string[]         // File extensions (default: ['.md', '.markdown', '.txt', '.mdx'])
}

`read_content_file`

Read and parse a single content file:

{
  filePath: string             // Path to file to read
}

🏗 Project Structure

content-manager-mcp/
├── src/
│   ├── types/          # Zod schemas and TypeScript types
│   ├── utils/          # Utility functions for file operations
│   │   ├── fileUtils.ts      # File system operations
│   │   ├── markdownUtils.ts  # Markdown processing
│   │   └── searchUtils.ts    # Search functionality
│   ├── tools/          # MCP tool definitions and handlers
│   └── index.ts        # Main server entry point
├── dist/               # Compiled output
├── tsup.config.ts      # Build configuration
├── tsconfig.json       # TypeScript configuration
└── package.json        # Project metadata and dependencies

🧪 Development

Scripts

pnpm dev - Development mode with watch
pnpm build - Build for production
pnpm start - Run built server
pnpm lint - Check code style
pnpm lint:fix - Fix code style issues
pnpm type-check - TypeScript type checking
pnpm clean - Clean build artifacts

Type Safety

This project leverages Zod for runtime validation combined with TypeScript for compile-time safety. All tool arguments are validated at runtime, ensuring robust error handling and type safety.

Error Handling

Comprehensive error catching and reporting
Zod validation errors with descriptive messages
File system error handling
Graceful degradation for malformed content

📝 Example Usage

Frontmatter Format

The server recognizes YAML frontmatter in Markdown files:

---
title: My Document
tags: [typescript, mcp, documentation]
author: Your Name
date: 2024-01-15
---

# Document Content

Your markdown content here...