pdfreaderMCP

# PDF Reader MCP Server An MCP server for reading PDF files and extracting text content. ## Installation ```bash cd /path/to/pdfreaderMCP pip install -r requirements.txt ``` ## MCP Configuration Add the following to your `~/.cursor/mcp.json` or `/root/.cursor/mcp.json`: ```json { "mcpServers": { "pdf-reader": { "command": "python3", "args": [ "/absolute/path/to/server.py" ] } } } ``` **Note:** Replace `/absolute/path/to/server.py` with the actual path to your `server.py` file. ## Available Tools ### 1. read_pdf Reads a PDF file and extracts text content. **Parameters:** - `file_path` (required): Path to the PDF file to read - `page_range` (optional): Page range to read - `"all"`: All pages (default) - `"1-5"`: Pages 1 to 5 - `"1,3,5"`: Pages 1, 3, and 5 only - `"1-3,7,10-12"`: Mixed range **Example:** ```python # Read all pages read_pdf(file_path="/path/to/document.pdf") # Read specific pages read_pdf(file_path="/path/to/document.pdf", page_range="1-10") ``` ### 2. get_pdf_info Retrieves PDF metadata information. **Parameters:** - `file_path` (required): Path to the PDF file **Example:** ```python get_pdf_info(file_path="/path/to/document.pdf") ``` ## Features - ✅ PDF text extraction - ✅ Page range specification - ✅ PDF metadata retrieval (title, author, dates, etc.) - ✅ Unicode support (Korean, Chinese, Japanese, etc.) ## Dependencies - `mcp>=1.0.0`: Model Context Protocol server library - `pymupdf>=1.24.0`: PDF processing library (PyMuPDF) ## Troubleshooting ### ImportError: mcp package not found ```bash pip install mcp ``` ### ImportError: fitz (PyMuPDF) not found ```bash pip install pymupdf ``` ## Restart Cursor After configuration changes, restart Cursor to activate the new MCP server. ## License MIT License - Free to use, modify, and distribute. See the [LICENSE](LICENSE) file for details.