CLDGeminiPDF MCP Server

What is CLDGeminiPDF?

CLDGeminiPDF is a Model Context Protocol (MCP) server that enables Claude Desktop to analyze PDF documents using Google's Gemini AI models. It sends PDF files and extracts text, leveraging Gemini’s language understanding for intelligent analysis and insights. It is intended for Claude Desktop users who want AI-powered PDF analysis.

How to use CLDGeminiPDF?

Download the pre-built JAR from the Releases page, obtain a free Gemini API key from Google AI Studio, and add the server to your Claude Desktop configuration file (claude_desktop_config.json) with the JAR path and API key. Restart Claude Desktop, then use commands like “Analyze this research paper: file:///path/to/paper.pdf” or “Find and analyze the file by name.”

Key features of CLDGeminiPDF

• PDF analysis using Gemini AI models
• Multiple model support (Gemini 2.5, 2.0, 1.5 series, and Gemma)
• Dual processing: direct PDF upload or fallback text extraction
• Seamless integration with Claude Desktop via MCP
• Environment-based configuration for easy deployment

Use cases of CLDGeminiPDF

• Analyze research papers with focus on methodology and conclusions
• Review contracts for key terms and potential risks using a specific Gemini model
• Find and analyze PDFs by filename within the allowed directory scope
• List available Gemini models and their capabilities

FAQ from CLDGeminiPDF

What prerequisites are needed to run CLDGeminiPDF?

Java 11 or higher, Maven (if building from source), a free Google AI Studio API key, the Claude Desktop application, and the Filesystem MCP server are required.

Where do I put the API key and model choice?

Set the GEMINI_API_KEY and GEMINI_MODEL environment variables in the env block of the CLDGeminiPDF server entry in Claude Desktop’s config file. The default model is gemini-2.0-flash.

What should I do if I see “Gemini API error: 401”?

Verify that your API key is correct and valid, and check that it has the necessary permissions in Google AI Studio.

Does version 1.0.0 support drag-and-drop of PDF files?

No. Version 1.0.0 does not support drag-and-drop files in the chat interface. Files must be accessible through the Filesystem MCP server.

Which Gemini models are available for free tier users?

Available models include gemini-2.5-flash-preview-05-20, gemini-2.5-pro-preview-05-06, gemini-2.0-flash, gemini-2.0-flash-lite, gemini-1.5-flash, and gemma-3 models. gemini-1.5-pro is unavailable for free tier users. Model availability depends on your API key and usage tier.