The GroupDocs.Search MCP server runs anywhere Docker runs. There is no NuGet channel for this product — The packed tool is 280.9 MiB — the GroupDocs.Search engine is about 200 MiB with its ONNX and SkiaSharp natives — which is over nuget.org’s 250 MB package limit, so dnx and dotnet tool install are not available for this product.
Docker channel (the only channel)
Runtime
Docker Desktop / Docker engine
.NET SDK
not required — the runtime is inside the image
Native dependencies
bundled in the image
Architectures
linux/amd64 + linux/arm64 (Apple Silicon native)
Image
about 380 MiB to download (382 MiB amd64, 378 MiB arm64 for 26.10.1)
Disk for indexes
grows with the extracted text of what you index; plan on a fraction of the documents’ size, more for text-heavy corpora
Network
pull from ghcr.io / docker.io at install only — none at runtime (metered licensing reports usage)
Linux
the container user must be able to write the index folder — --user "$(id -u):$(id -g)"
Pull the image deliberately (docker pull ghcr.io/groupdocs-search/search-net-mcp:latest) rather than letting a client hit it on the first tool call and time out.
AI client requirements: any MCP-compatible client (Claude Desktop, Claude Code, VS Code / GitHub Copilot, Visual Studio 2022 17.14+, Cursor, Windsurf, Cline, Codex CLI) — see Register in AI clients.