GroupDocs.Conversion MCP server lets AI agents like Claude, Cursor, and Copilot convert documents between 100+ formats — PDF, Word, Excel, PowerPoint, images, Markdown — locally on your machine. Files are never uploaded to any cloud service. Install with one command:
dnx GroupDocs.Conversion.Mcp --yes
Or use the guided installer to register the server in your AI client, verify the setup, and configure shared folders in one pass.
What you can do
The server exposes three tools to any MCP-compatible agent (full details in the tools reference):
convert — convert a document to another format (PDF, DOCX, XLSX, PPTX, HTML, PNG, JPG, Markdown, and many more) and save it to your storage folder.
get_supported_formats — list every target format a given document can be converted to.
get_document_info — file type, page count, and basic properties, without converting.
Ask your agent in plain language — “Convert report.docx to PDF”, “Turn this PDF into Markdown” — and it picks the right tool.
Conversion vs extraction
This server does format conversion: full-fidelity transformation of a document into another format (LLM-ready Markdown included), preserving layout, headings, tables, and lists. If you need field-level data extraction — pulling specific values, structured tables, or text spans out of documents — use the companion GroupDocs.Parser MCP server. Many pipelines use both: convert for ingestion, parse for extraction.
Supported AI clients
Client
How it connects
Claude Desktop
claude_desktop_config.json
Claude Code
claude mcp add CLI
VS Code / GitHub Copilot
user-level or workspace mcp.json
Visual Studio 2022 (17.14+)
.mcp.json in the solution root
Cursor
~/.cursor/mcp.json
Windsurf
~/.codeium/windsurf/mcp_config.json
Cline
Cline MCP settings
Codex CLI
codex mcp add CLI
JetBrains Rider
manual registration (Settings → AI Assistant → MCP)
The server uses MCP’s local stdio transport: your AI client starts the server as a child process and talks to it over standard input/output. There are no inbound ports, no external endpoints, and no telemetry — the data path is agent → local server → local filesystem. That makes it suitable for regulated and internal documents that must not leave your machine. Details: On-premise architecture.
When you need more than a basic Markdown converter
Simple Markdown-extraction servers cover plain text well. Choose this server when you need: complex layouts and tables preserved, Office formats in both directions (DOCX/XLSX/PPTX as output, not only input), PDF/A for archiving, image renditions (PNG/JPG previews), password-protected documents, and the rendering fidelity of the commercial GroupDocs engine trusted by enterprise teams for over a decade.