home / mcp / mcp content engineering mcp server

MCP Content Engineering MCP Server

Module for configuring and managing content pipelines with optional Markdown support.

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "hendrickcastro-mcpcontentengineering": {
      "command": "npx",
      "args": [
        "-y",
        "hendrickcastro/MCPContentEngineering"
      ],
      "env": {
        "CONTENT_SOURCE_PATH": "/path/to/your/docs/",
        "CONTENT_SOURCE_TYPE": "directory"
      }
    }
  }
}

This MCP server provides direct access to raw Markdown content by serving Markdown files exactly as written. You can point it at a single file or an entire directory of .md files, and the server will return the content with metadata that helps you understand what was read. This is ideal for feeding knowledge bases, architecture docs, or coding standards to AI models without any parsing or transformation overhead.

How to use

Attach an MCP client and load the MCP server configuration you’ve prepared. You will be able to request raw Markdown content from a single file or merge all Markdown files found in a directory, with clear separators identifying each source file. You’ll receive the exact text of the Markdown files, plus metadata that describes what was read and where it came from.

How to install

Prerequisites you need before you begin:

Node.js 18+ with npm
Access to Markdown files with a .md extension
An MCP-compatible client such as Claude Desktop or Cursor IDE

# Option 1: Run via npx from a GitHub-hosted MCP server (recommended for quick testing)
# Claude Desktop example configuration uses a JSON snippet with environment vars
# Cursor IDE example configuration uses a JSON snippet with environment vars

# Example for Claude Desktop (mcp_servers.json):
{
  "mcpServers": {
    "mcp-content-engineering": {
      "command": "npx",
      "args": ["-y", "hendrickcastro/MCPContentEngineering"],
      "env": {
        "CONTENT_SOURCE_TYPE": "file",
        "CONTENT_SOURCE_PATH": "/path/to/your/business-rules.md"
      }
    }
  }
}
```} ,{

Configuration and usage notes

Two environment variables control what content you read. Use one of these configurations to point at a single Markdown file or at a directory containing multiple Markdown files.

{
  "mcpServers": {
    "mcp-content-engineering": {
      "command": "npx",
      "args": ["-y", "hendrickcastro/MCPContentEngineering"],
      "env": {
        "CONTENT_SOURCE_TYPE": "directory",
        "CONTENT_SOURCE_PATH": "/path/to/your/docs/"
      }
    }
  }
}

Usage patterns and expected outputs

Single File Access returns the exact content of a single Markdown file, including file size and last modification details.

Directory Mode combines all Markdown files found under the specified directory, inserting clear separators between files to preserve source boundaries.

Available tools

content_get_raw

Raw Content Access tool that returns Markdown content without processing, including metadata about source and size.