MCP OCR Server

A Model Context Protocol (MCP) server that provides OCR (Optical Character Recognition) capabilities for images and videos using GLM-4.1V-Thinking-Flash.

Features

• Image OCR: Extract text from images using GLM-4.1V-Thinking-Flash
• Video OCR: Extract text from video frames (planned feature)
• Custom Prompts: Support for custom OCR prompts
• MCP Integration: Full MCP protocol support for seamless integration

Requirements

• Python 3.10+
• GLM-4.1V-Thinking-Flash API key from ZhipuAI

Quick Start with uvx

The fastest way to get started is using uvx:

# Install and run directly
uvx mcp-server-ocr

# Or install for development
uvx --python 3.10 --with-editable . mcp-server-ocr

Installation

Using uv (recommended)

# Clone the repository
git clone <repository-url>
cd ocr

# Install dependencies
uv sync

# Run the server
uv run mcp-server-ocr

Using pip

# Clone the repository
git clone <repository-url>
cd ocr

# Install dependencies
pip install -e .

# Run the server
mcp-server-ocr

Configuration

Set your GLM-4.1V-Thinking-Flash API key as an environment variable:

export ZHIPU_API_KEY="your-api-key-here"

Usage

Available Tools

1. ocr_image: Perform OCR on image files

   • image_path: Path to the image file
   • prompt: Custom prompt for OCR processing (optional)

2. ocr_video: Perform OCR on video frames (coming soon)

   • video_path: Path to the video file
   • prompt: Custom prompt for video OCR processing (optional)
   • frame_interval: Extract frames every N seconds (optional)

Available Prompts

1. ocr_image: Extract text from an image
2. ocr_video: Extract text from video frames

Build and Publish

Build the package

# Using uv
uv build

# Using pip
python -m build

Publish to PyPI

# Using uv
uv publish

# Using twine
twine upload dist/*

Development

Setup development environment

# Clone and setup
git checkout -b feature/your-feature
uv sync --dev

# Run linting and type checking
uv run ruff check .
uv run pyright .

Testing

# Run tests (when available)
uv run pytest

# Test the server manually
uv run python -m mcp_server_ocr

API Documentation

• GLM-4.1V-Thinking-Flash API Documentation
• GLM-4.1V-Thinking Usage Guide

MCP Protocol

This server implements the Model Context Protocol (MCP) specification:

• Python MCP SDK
• TypeScript MCP SDK

License

MIT License