MCP Document Converter

What is MCP Document Converter?

MCP Document Converter is a lightweight Model Context Protocol server that converts documents to Markdown. It supports PDF, DOCX, HTML, EPUB, CSV, JSON, plain text, and images (via OCR). The server is built for AI agents that need to ingest and understand document content.

How to use MCP Document Converter?

Clone the repository, install dependencies with pip install -r requirements.txt, and run python -m mcp_server_convert. Configure in Claude Code by adding a conversion server entry to your MCP settings. A Docker image is also provided for containerized deployment.

Key features of MCP Document Converter

• Multi-format support: PDF, DOCX, HTML, EPUB, CSV, JSON, images, and more
• Six MCP tools: convert_file, convert_url, batch_convert, and others
• Zero external dependencies for core conversion (uses Python standard library)
• Fast in-memory processing with no temporary files
• Docker-ready with a single Dockerfile and one-command deployment

Use cases of MCP Document Converter

• Convert a PDF or DOCX document to Markdown for LLM ingestion
• Fetch a URL and convert its HTML content to Markdown
• Batch convert multiple files or an entire directory of documents
• Extract metadata from documents without full conversion

FAQ from MCP Document Converter

What file formats does MCP Document Converter support?

It supports PDF, DOCX, HTML, EPUB, CSV, JSON, XML, Excel (XLSX), PowerPoint (PPTX), plain text (TXT, MD, RST, LOG), and images (PNG, JPG via OCR).

How do I run MCP Document Converter with Docker?

Build the image with docker build -t mcp-server-convert . then run docker run -i --rm mcp-server-convert. For persistent file access, mount a volume with the -v flag.

How do I configure MCP Document Converter in Claude Code?

Add a "convert" entry to your MCP settings in ~/.claude/settings.json, pointing the command to python -m mcp_server_convert with the working directory set to the project path.

What are the main tools provided?

The server offers six tools: convert_file, convert_url, batch_convert, convert_directory, extract_metadata, and list_supported_formats.

Does MCP Document Converter require external dependencies?

Core conversion uses Python standard library and markdownify for HTML. Additional formats like PDF, DOCX, and Excel require optional libraries (PyMuPDF, python-docx, openpyxl) listed in requirements.txt.