# Running GroupDocs MCP servers on-premise: architecture and security model Run metadata processing for AI agents fully on-premise: local stdio transport, no external endpoints, no inbound ports, no telemetry. Run metadata reading, editing, and removal for AI agents **fully on-premise**: the GroupDocs.Metadata MCP server uses local stdio transport with **no external endpoints, no inbound ports, and no telemetry**. This page is the one to send your security reviewer. {{< alert style="info" >}} The commands and config snippets on this page are for the **.NET** build of the server — the only platform available today. Installation and client setup: [MCP server for .NET]({{< ref "metadata/net/mcp/_index.md" >}}). Other platforms will expose the same tools with their own launch command; everything else on this page applies unchanged. {{< /alert >}} ## The architecture in one picture ```text +--------------+ +--------------------+ +------------------+ | AI client | stdio | MCP server process | reads / | local filesystem | | (Claude, VS | <-----> | (GroupDocs engine) | <-----> | storage / output | | Code, agent) | JSON-RPC | child process | writes | folders | +--------------+ +--------------------+ +------------------+ ``` * **Transport:** the AI client *starts the server as a child process* and communicates over standard input/output. The server never listens on a network socket. * **Data path:** agent → local server → local filesystem. Files are read and written in the folders you configure; no file content is transmitted anywhere. * **Network use:** only at install time (nuget.org or ghcr.io/docker.io). At runtime the server makes no outbound calls. Air-gapped: pre-pull the image or pre-cache the package and pin the version. * **Telemetry:** none. The engine processes files in-process. ## Why this product in particular belongs on-premise Metadata is where the sensitive residue lives: author names, machine and network paths, company fields inherited from templates, GPS coordinates in photographs. Sending a file to a cloud service **to find out whether it is safe to share** is self-defeating — the service now has the file and everything in it. Here the file is read locally. What travels is the conversation: the properties the agent reports back to you. With a cloud-hosted model, "the author is Jane Smith and the photo was taken at these coordinates" reaches the model provider — which for a privacy review may be precisely the data you were protecting. Pair with a locally-hosted model and nothing leaves at all. ## Docker deployment inside the perimeter ```bash docker run --rm -i --user "$(id -u):$(id -g)" \ -v /srv/files:/data \ -v /srv/licenses:/license:ro \ -e GROUPDOCS_MCP_STORAGE_PATH=/data \ -e GROUPDOCS_LICENSE_PATH=/license/GroupDocs.Metadata.lic \ ghcr.io/groupdocs-metadata/metadata-net-mcp:26.9.0 ``` * Pin the tag (`:26.9.0`, not `:latest`). * `--user "$(id -u):$(id -g)"` (Linux) lets the container, which runs as a non-root user, write into a folder you own. A JSON client config runs no shell to expand it — write the numbers: `"--user", "1000:1000"`. * Licence read-only; mount only the folders the agent should reach. * Multi-arch images (linux/amd64 + linux/arm64) with all native dependencies included. ## License management * **License file** — read from local disk by the local process. Fully offline; the right answer for air-gapped deployments, and the only way to get complete metadata reads. * **Metered (pay-per-use)** — reports *usage* to GroupDocs servers, so it needs outbound egress. File content is never part of that report. Both are covered in [Licensing]({{< ref "metadata/mcp/getting-started/licensing.md" >}}). ## What this fits — honestly **A good fit:** pre-publication privacy checks, e-discovery preparation, bulk property correction after migrations, and any environment where files cannot be uploaded for inspection. **Not what this is:** a DLP product or a scanning service. It answers questions about the files you point it at, one client on one machine — it does not watch a share, enforce a policy, or quarantine anything. ## FAQ **Does any file content leave the machine?** Not from the server. Properties the agent reports back travel in the conversation to your model provider. **Does it need internet at runtime?** No — only at install, and when metered licensing is enabled. **Can I run it air-gapped?** Yes: pre-pull the image, use a license file, pin the version. **What ports does it open?** None. stdio only. **How do I prove that?** The [verification script]({{< ref "metadata/net/mcp/troubleshooting.md" >}}#verifying-an-installation-end-to-end) performs a real handshake and a real engine call, so you can watch exactly what happens. ## Read more * [3 architectures for AI document processing, and the one that keeps files inside your network](https://blog.groupdocs.com/total/on-premise-ai-document-processing-mcp/) — the same scenario on the GroupDocs blog, with worked examples and the trade-offs.