GroupDocs.Metadata MCP Server Leave feedback

GroupDocs.Metadata MCP server lets AI agents like Claude, Cursor, and Copilot read, edit, and strip metadata from documents and images — PDF, Office, images, audio, video and more — locally on your machine. The files stay where they are; only the properties change.

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-metadata/metadata-net-mcp:latest

With the .NET 10 SDK installed, the same server also runs without Docker:

dnx GroupDocs.Metadata.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.

Six tools (full details in the tools reference):

• read_metadata — every property in the file as JSON: author, title, dates, custom fields, EXIF, XMP, IPTC.
• search_metadata — only the properties matching a category, name, or value filter.
• write_metadata — set, change, or add one property and save the updated file.
• remove_metadata — strip everything, or just the categories you name.
• get_document_info — type, size, pages, encryption status, without enumerating properties.
• get_license_status — which licensing mode is active and, under metered, how much has been consumed.

Ask in plain language — “who wrote this?”, “strip everything before I send it” — and the agent picks the tools.

Installation, prerequisites, and client configuration are platform-specific; the tools and licensing model below are the same everywhere.

Warning

Read this before trusting a metadata audit. In evaluation mode the engine returns only the first five document properties, with reduced XMP and EXIF access — and nothing in the response says so. An unlicensed privacy check can report a clean file that is not clean. Confirm with get_license_status first; see Licensing.

Exact config blocks for every client: Register in AI clients.

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. No inbound ports, no external endpoints, no telemetry — the data path is agent → local server → local filesystem. Metadata is exactly the kind of data you do not want to hand to a cloud service to inspect: it is where author names, machine names, GPS coordinates, and internal paths hide. Details: On-premise architecture.

Every application shows a few properties for its own format. Choose this server when you need: one model across documents, images, audio, and video; the packages a UI hides (XMP, IPTC, EXIF including GPS); value search across every package at once; bulk edit and strip driven from a prompt rather than by hand; and the fidelity of the commercial GroupDocs engine trusted by enterprise teams for over a decade.

• Quick start · Use cases · Troubleshooting & FAQ
• GitHub: server source · installer · integration tests
• NuGet package · Docker image · MCP Registry
• Questions: Metadata forum