# Cursor MCP Configuration
Integration guide for Cursor IDE with OCR PDF MCP server.
## Prerequisites
1. Install Python dependencies:
```bash
pip install -r requirements.txt
```
2. Ensure Tesseract is installed and in PATH
## Cursor Configuration
### Method 1: VS Code Extension Style
Create `.cursor/extensions.json`:
```json
{
"mcpServers": {
"ocr-pdf": {
"command": "python",
"args": [
"D:\\AI\\MCP\\python\\ocr_pdf_mcp\\run.py"
],
"env": {
"PYTHONPATH": "D:\\AI\\MCP\\python\\ocr_pdf_mcp"
}
}
}
}
```
### Method 2: Settings Configuration
Add to Cursor settings.json:
```json
{
"mcp.servers": {
"ocr-pdf": {
"command": "python",
"args": ["D:\\AI\\MCP\\python\\ocr_pdf_mcp\\run.py"],
"env": {
"PYTHONPATH": "D:\\AI\\MCP\\python\\ocr_pdf_mcp",
"MCP_SERVER_HOST": "localhost",
"MCP_SERVER_PORT": "8000"
}
}
}
}
```
### Method 3: HTTP Server Integration
1. Start HTTP server:
```bash
python run.py
```
2. Configure Cursor for HTTP:
```json
{
"mcp.httpServers": {
"ocr-pdf": {
"url": "http://localhost:8000",
"endpoints": {
"initialize": "/mcp/initialize",
"tools/list": "/mcp/tools/list",
"tools/call": "/mcp/tools/call"
}
}
}
}
```
## Usage in Cursor
### Command Palette
1. Open Command Palette (`Ctrl+Shift+P`)
2. Search for "MCP: OCR PDF"
3. Select desired tool:
- Smart PDF Processing
- Text Extraction
- OCR Image
- PDF Analysis
- Batch Processing
### Chat Integration
Use the following prompts in Cursor Chat:
```
@ocr-pdf Process this PDF file: /path/to/document.pdf
@ocr-pdf Extract text from: /path/to/scanned.pdf
@ocr-pdf Analyze structure of: /path/to/complex.pdf
```
### Inline Code Actions
Right-click on PDF files in explorer → "Process with OCR PDF MCP"
## Tool Specifications
### Smart PDF Processing
- **Command**: `process_pdf_smart`
- **Input**: PDF file path, language (optional)
- **Output**: Extracted text with metadata
### Text Extraction
- **Command**: `extract_pdf_text`
- **Input**: PDF file path, page range (optional)
- **Output**: Raw text content
### OCR Processing
- **Command**: `perform_ocr`
- **Input**: Image file path, language (optional)
- **Output**: OCR text results
### PDF Analysis
- **Command**: `analyze_pdf_structure`
- **Input**: PDF file path
- **Output**: Document structure and metadata
### Batch Processing
- **Command**: `batch_process_pdfs`
- **Input**: Directory path, output directory, file pattern
- **Output**: Processing results summary
## Keybindings
Add to `keybindings.json`:
```json
[
{
"key": "ctrl+shift+p",
"command": "mcp.ocr-pdf.process-smart",
"when": "resourceExtname == .pdf"
},
{
"key": "ctrl+shift+o",
"command": "mcp.ocr-pdf.perform-ocr",
"when": "resourceExtname =~ /\\.(png|jpg|jpeg|tiff|bmp)$/"
}
]
```
## Workspace Configuration
For project-specific configuration, create `.cursor/workspace.json`:
```json
{
"folders": [
{
"path": "."
}
],
"extensions": {
"recommendations": ["mcp-ocr-pdf"]
},
"settings": {
"mcp.ocr-pdf.defaultLanguage": "eng",
"mcp.ocr-pdf.outputFormat": "text",
"mcp.ocr-pdf.autoStart": true
}
}
```
## Troubleshooting
### Server Not Starting
```bash
# Check if server starts manually
python run.py
```
### Permission Issues
- Run Cursor as administrator
- Check file path permissions
- Verify Python executable path
### OCR Issues
- Install Tesseract: `choco install tesseract` (Windows)
- Add Tesseract to PATH
- Check language data files
### Connection Problems
- Verify port 8000 is not in use
- Check Windows Firewall settings
- Test HTTP endpoint: `curl http://localhost:8000/health`
## Development
For development with debug mode:
```bash
python run_server.py --dev --host 127.0.0.1 --port 8000
```
Monitor logs in Cursor Output panel under "MCP OCR PDF" channel.
## Repository
GitHub: https://github.com/irev/mcp-readpdfx
