Skip to content

MCP.md

Latest commit

 

History

History
109 lines (77 loc) · 4.15 KB

MCP.md

File metadata and controls

109 lines (77 loc) · 4.15 KB

MCP Document Processor

An intelligent document processing system that uses the Model Context Protocol (MCP) to extract, analyze, and route business documents automatically.

Overview

The MCP Document Processor is a full-stack application (FastAPI backend + React frontend) that automates document processing workflows. The system classifies documents, extracts relevant information, and provides a user-friendly interface for document management.

How MCP Works

The Model Context Protocol (MCP) is the core framework that powers this document processor. Here's how it works in simple terms:

  1. Document Context: When a document is uploaded, MCP creates a "context" object that acts as a container for:

    • The original document text
    • Document metadata (filename, upload time, etc.)
    • Processing history log
    • Extracted information with confidence scores

  2. Memory System: These context objects are stored in a persistent memory system (file-based storage), allowing them to survive between API calls and server restarts.

  3. Smart Routing: The system examines each document and routes it to the appropriate specialized processor:

    Document → Router → Invoice/Contract/Email/Default Processor

  4. Processing Pipeline: Each document follows this workflow:

    Upload → Classification → Information Extraction → Storage → Viewing

  5. History & Confidence Tracking: Every processing step is recorded with timestamps and confidence scores, providing full transparency into how information was extracted.

Key Features

  • Document Classification: Automatically identifies document types (invoices, contracts, emails)
  • Information Extraction: Extracts key data using ML models with confidence scoring
  • Persistent Storage: File-based document storage that persists between server restarts
  • Processing Pipeline: Specialized processors for different document types
  • Modern UI: React-based dashboard for document upload, viewing, and management

Architecture

Backend

  • Context System: MCPContext objects maintain document state throughout processing
  • Memory System: Configurable storage (in-memory or file-based) with TTL management
  • Processor Router: Routes documents to specialized processors based on content
  • Document Processors: Type-specific processors with a default fallback processor
  • ML Models: Mock implementations of entity extraction, classification, and summarization

Frontend

  • Dashboard: Document statistics and listing with filtering options
  • Document View: Detailed view of processed documents and extracted data
  • Upload Interface: Simple drag-and-drop document upload

Setup

Backend

# Create and activate virtual environment
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

# Start server
python app.py  # Runs on http://localhost:8000

Frontend

# Navigate to frontend directory
cd frontend

# Install dependencies
npm install

# Start development server
npm start  # Runs on http://localhost:3000

API Endpoints

  • POST /api/documents/upload: Upload a new document
  • POST /api/documents/{document_id}/process: Process a document
  • GET /api/documents/{document_id}: Get document details
  • GET /api/documents: List all documents
  • DELETE /api/documents/{document_id}: Delete a document
  • GET /api/memory-status: Check memory store status (debugging)

UI Features

  • Dashboard: Document statistics and listing with filtering options Dashboard UI
  • Upload Interface: Simple drag-and-drop document upload Upload Interface UI
  • Document View: Detailed view of processed documents and extracted data Document View UI

Technical Notes

  • Documents are stored in data/documents/ with a 24-hour TTL
  • The DefaultProcessor handles any document type that specialized processors cannot handle
  • CORS is enabled to allow frontend-backend communication
  • Proxy is configured in package.json to route API requests during development