Claudehopper (Construction Document Analysis)

🚀 ClaudeHopper - AI-Powered Construction Document Assistant

ClaudeHopper is a professional Model Context Protocol (MCP) server that enables direct interaction between Claude and other Large Language Models (LLMs) with construction drawings, documents, and specifications. Through advanced Retrieval-Augmented Generation (RAG) and hybrid search capabilities, you can easily ask questions about construction drawings, find specific details, and analyze technical specifications.

✨ Features

• 🔍 Vector-based search optimized for construction drawings, plans, and specifications.
• 🖼️ Image retrieval based on text descriptions for finding similar drawings.
• 🏢 Metadata extraction tailored to construction industry document formats.
• 📊 Efficient token usage through intelligent document chunking.
• 🔒 Security ensured by local storage and processing.
• 📈 Support for multiple drawing types and construction disciplines (structural, civil, architectural, etc.).

🚀 Quick Start

Prerequisites

• Node.js 18+
• Ollama for local AI models.
  • Required models: nomic-embed-text, phi4, clip.
• Claude desktop application.
• Image extraction (optional): Poppler Utils (pdfimages command).

One-click Setup

1. Download ClaudeHopper.
2. Run the installation script:

cd ~/Desktop/claudehopper
chmod +x run_now_preserve.sh
./run_now_preserve.sh

This will:

• Create the necessary directory structure.
• Install the required AI models.
• Process your construction drawings.
• Configure the Claude desktop application to use ClaudeHopper.

Adding Documents

Place your construction documents in the following folders:

• Drawings: ~/Desktop/PDFdrawings-MCP/InputDocs/Drawings/
• Specifications: ~/Desktop/PDFdrawings-MCP/InputDocs/TextDocs/

After adding documents, run:

./process_pdfdrawings.sh

💻 Usage Examples

Using ClaudeHopper with Claude

Try the following sample questions in the Claude desktop application:

“What construction drawings do we have for this project?”
“Show the structural details of the foundation system.”
“Find the drawings showing the concrete foundation and its dimensions.”
“Search for the lift station layout drawing.”
“What are the specifications for Interior paint?”
“Find all content discussing the fire protection system.”

🔧 Technical Details

Technical Architecture

ClaudeHopper uses a multi-stage pipeline to process construction documents:

1. Document Analysis: Analyze the structure and content type of PDF documents.
2. Metadata Extraction: Extract project information, drawing types, and disciplines from documents with AI assistance.
3. Content Chunking: Intelligently split documents to maintain context.
4. Image Extraction: Identify and extract drawing images from PDFs.
5. Vector Embedding: Create semantic representations of text and images.
6. Database Storage: Store data in the database.

Database Description

• Table Structure:
  • documents table: Stores basic document information (e.g., file name, content summary, etc.).
  • metadata table: Stores extracted metadata information (e.g., author, date, etc.).
  • vectors table: Stores vector representations for similarity search.

📄 License

This project is licensed under the MIT License. See the LICENSE file for details.

⚠️ Important Note

Ensure that all necessary dependencies, including Node.js and Ollama, are installed.

💡 Usage Tip

When processing a large number of documents, it is recommended to optimize the hardware configuration to improve performance. For further support or customization, refer to the project documentation or contact the maintenance team.

Welcome to use ClaudeHopper! For more help, please feel free to check our documentation or contact us.