GroupDocs.Comparison MCP server lets AI agents like Claude, Cursor, and Copilot compare two versions of a document — Word, PDF, Excel, PowerPoint, OpenDocument, text, HTML — locally on your machine, and either explain what changed or hand you a marked-up file. Nothing is uploaded to any cloud service.
Run it with one command. The Docker image is self-contained — the runtime and every native dependency the engine needs are inside it:
docker run --rm -i -v $(pwd)/documents:/data \
ghcr.io/groupdocs-comparison/comparison-net-mcp:latest
With the .NET 10 SDK installed, the same server also runs without Docker:
dnx GroupDocs.Comparison.Mcp --yes
Both are the .NET build of the server and run on Windows, Linux, and macOS. Other platforms will each get their own launcher — see Install for your platform.
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 four tools to any MCP-compatible agent (full details in the tools reference):
compare — compare two documents and save a marked-up result file with every difference highlighted.
analyze_changes — get the differences as structured data without rendering a file: faster, cheaper, and what the agent needs to answer “what changed?”.
get_document_info — file type, page count, size, and per-page dimensions, without comparing.
get_license_status — which licensing mode is active (evaluation, license file, or metered) and, under metered, how much has been consumed.
Ask your agent in plain language — “What changed in the new contract?”, “Give me a redline of these two decks” — and it picks the right tool.
Install for your platform
Installation, prerequisites, and client configuration are platform-specific; the tools and licensing model below are the same everywhere.
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 matters more here than in most integrations: the documents people compare are contracts, filings, and drafts under negotiation. Details: On-premise architecture.
When you need more than a text diff
diff and track-changes cover plain text and same-application edits. Choose this server when you need: format-aware comparison across PDF, Word, Excel, PowerPoint and OpenDocument; changes attributed to the right page, paragraph, and table cell; style changes (bold, font, colour) detected as changes rather than ignored; a marked-up document your reviewers can open in their own tools; password-protected files; and the fidelity of the commercial GroupDocs engine trusted by enterprise teams for over a decade.