Documentation Management System MCP server for AI agents

The DocMCP server provides a powerful way to index, store, and query documentation for large language models. Built on PostgreSQL with pgvector, it crawls documentation sites, processes content, generates embeddings, and makes the information accessible through AI IDEs like Cursor.

Installation

Prerequisites

Docker and Docker Compose
Node.js 16+
Git
AWS Account with Bedrock access
AWS CLI with appropriate credentials

Setup Instructions

Clone the repository

git clone https://github.com/visheshd/docmcp.git
cd docmcp

Configure environment variables

cp .env.example .env

Edit the .env file with your settings:

DATABASE_URL=postgresql://postgres:postgres@localhost:5433/docmcp
AWS_REGION=your-aws-region
AWS_ACCESS_KEY_ID=your-access-key
AWS_SECRET_ACCESS_KEY=your-secret-key
LOG_LEVEL=info

Start the development server
```
chmod +x dev-start.sh
./dev-start.sh
```
This starts PostgreSQL with pgvector, installs dependencies, runs migrations, and imports seed data.

Usage

Adding Documentation

Use the add-docs script to crawl and process documentation:

# Basic usage
npm run add-docs -- --url https://example.com/docs --max-depth 3

# Advanced usage with additional options
npm run add-docs -- \
  --url https://example.com/docs \
  --max-depth 3 \
  --tags react,frontend \
  --package react \
  --version 18.0.0 \
  --wait

Available Options

--url: Documentation URL to crawl (required)
--max-depth: Maximum crawl depth (default: 3)
--tags: Comma-separated tags for categorization
--package: Package name this documentation is for
--version: Package version (defaults to "latest")
--wait: Wait for processing to complete
--verbose: Enable detailed logging

For a complete list of options:

npm run add-docs -- --help

Querying Documentation

After adding documentation, you can query it using the MCP tools integrated into your AI IDE.

Stopping the Server

To stop the development environment:

docker-compose -f docker-compose.dev.yml down

Cursor IDE Integration

To use DocMCP with Cursor IDE, add the following configuration to your Cursor settings:

{
    "docmcp-local-stdio": {
      "transport": "stdio",
      "command": "node",
      "args": [
        "/absolute/path/to/docmcp/dist/stdio-server.js"
      ],
      "clientInfo": {
        "name": "cursor-client",
        "version": "1.0.0"
      }
    }
}

Replace /absolute/path/to/docmcp with the actual path to your DocMCP installation.

After adding this configuration, restart Cursor for the changes to take effect.

How It Works

Document Processing Pipeline

DocMCP processes documentation through these steps:

Documentation Input: User provides a URL and system creates a job record
Web Crawling: Crawler respects robots.txt restrictions and follows links to specified depth
Document Processing: HTML is cleaned and converted to structured Markdown
Chunking & Embedding: Documents are split into chunks and vector embeddings are generated
Job Finalization: Job status is updated and documents become available for querying
Querying & Retrieval: Users can query the system to find relevant documentation chunks

The system leverages AWS Bedrock for generating vector embeddings, enabling powerful semantic search capabilities for documentation content.

How to install this MCP server

For Claude Code

To add this MCP server to Claude Code, run this command in your terminal:

claude mcp add-json "docmcp-local-stdio" '{"transport":"stdio","command":"node","args":["<DOCMCP_DIR>/dist/stdio-server.js"],"clientInfo":{"name":"cursor-client","version":"1.0.0"}}'

See the official Claude Code MCP documentation for more details.

For Cursor

There are two ways to add an MCP server to Cursor. The most common way is to add the server globally in the ~/.cursor/mcp.json file so that it is available in all of your projects.

If you only need the server in a single project, you can add it to the project instead by creating or adding it to the .cursor/mcp.json file.

Adding an MCP server to Cursor globally

To add a global MCP server go to Cursor Settings > Tools & Integrations and click "New MCP Server".

When you click that button the ~/.cursor/mcp.json file will be opened and you can add your server like this:

{
    "mcpServers": {
        "docmcp-local-stdio": {
            "transport": "stdio",
            "command": "node",
            "args": [
                "<DOCMCP_DIR>/dist/stdio-server.js"
            ],
            "clientInfo": {
                "name": "cursor-client",
                "version": "1.0.0"
            }
        }
    }
}

Adding an MCP server to a project

To add an MCP server to a project you can create a new .cursor/mcp.json file or add it to the existing one. This will look exactly the same as the global MCP server example above.

How to use the MCP server

Once the server is installed, you might need to head back to Settings > MCP and click the refresh button.

The Cursor agent will then be able to see the available tools the added MCP server has available and will call them when it needs to.

You can also explicitly ask the agent to use the tool by mentioning the tool name and describing what the function does.

For Claude Desktop

To add this MCP server to Claude Desktop:

1. Find your configuration file:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
Linux: ~/.config/Claude/claude_desktop_config.json

2. Add this to your configuration file:

{
    "mcpServers": {
        "docmcp-local-stdio": {
            "transport": "stdio",
            "command": "node",
            "args": [
                "<DOCMCP_DIR>/dist/stdio-server.js"
            ],
            "clientInfo": {
                "name": "cursor-client",
                "version": "1.0.0"
            }
        }
    }
}

3. Restart Claude Desktop for the changes to take effect