Skip to main content

Installation

Get Docsray MCP up and running in minutes.

Requirements​

  • Python 3.9 or higher
  • pip package manager
  • Claude Desktop or compatible MCP client

Install via pip​

pip install docsray-mcp

Install from source​

git clone https://github.com/docsray/docsray-mcp.git
cd docsray-mcp
pip install -e .

Provider Setup​

PyMuPDF (Default - No Setup Required)​

PyMuPDF4LLM is included and works out of the box:

# Already included, no action needed!

LlamaParse (AI-Powered Analysis)​

For comprehensive extraction with AI analysis:

  1. Get your API key from LlamaIndex Cloud
  2. Set the environment variable:
export LLAMAPARSE_API_KEY="your-api-key-here"

Or add to your .env file:

LLAMAPARSE_API_KEY=your-api-key-here

Configure Claude Desktop​

Add Docsray to your Claude Desktop configuration:

macOS​

Edit ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "docsray": {
      "command": "uvx",
      "args": ["docsray-mcp"],
      "env": {
        "LLAMAPARSE_API_KEY": "your-api-key-here"
      }
    }
  }
}

Windows​

Edit %APPDATA%\Claude\claude_desktop_config.json:

{
  "mcpServers": {
    "docsray": {
      "command": "uvx",
      "args": ["docsray-mcp"],
      "env": {
        "LLAMAPARSE_API_KEY": "your-api-key-here"
      }
    }
  }
}

Alternative: Using pip installation​

If you installed with pip and want to use the Python module directly:

{
  "mcpServers": {
    "docsray": {
      "command": "docsray",
      "args": [],
      "env": {
        "LLAMAPARSE_API_KEY": "your-api-key-here"
      }
    }
  }
}

Verify Installation​

Test with Claude:

You: Peek at https://www.w3.org/WAI/ER/tests/xhtml/testfiles/resources/pdf/dummy.pdf

Claude: [Uses Docsray to analyze the PDF and show its structure]

Optional Configuration​

Environment Variables​

# LlamaParse configuration
LLAMAPARSE_API_KEY=your-api-key
LLAMAPARSE_MODE=fast  # or 'accurate', 'premium'

# Performance tuning
DOCSRAY_MAX_CONCURRENT_REQUESTS=5
DOCSRAY_TIMEOUT_SECONDS=30

# Caching
DOCSRAY_CACHE_DIR=.docsray  # Custom cache directory

Provider Selection​

In Claude or your MCP client, you can specify providers:

# Force specific provider
"Xray document.pdf with provider llama-parse"
"Peek at document.pdf using provider pymupdf4llm"

# Let Docsray choose (default)
"Xray document.pdf"  # Auto-selects best provider

Troubleshooting​

LlamaParse not working?​

  • Check your API key is set correctly
  • Verify you have API credits available
  • Try the fast mode: LLAMAPARSE_MODE=fast

PyMuPDF issues?​

  • Ensure Python 3.9+ is installed
  • Try reinstalling: pip install --upgrade pymupdf4llm

Cache issues?​

  • Clear cache: rm -rf .docsray/
  • Check disk space availability

Next Steps​

Now that you have Docsray installed, check out the Quickstart Guide to start extracting data from documents!