GroupDocs.Viewer Cloud MCP Server

(Docker image uses Python 3.12)

A lightweight MCP (Model Context Protocol) server that wraps GroupDocs.Viewer Cloud so AI agents can inspect supported formats, read document page info, render views (HTML/PNG/JPG/PDF) to cloud storage, and use shared storage utilities.

---

Features

• Viewer: supported formats, document info (pages), create view, delete view output
• Storage: upload, download, list, exists, delete (same pattern as other GroupDocs Cloud MCP servers)
• Cross-platform: Windows, macOS, Linux

---

Quick start

1. Clone the published repository

git clone git@github.com:groupdocs-viewer-cloud/groupdocs-viewer-cloud-mcp.git
cd groupdocs-viewer-cloud-mcp

2. Configure environment variables

Create a .env file (or copy .env.example):

CLIENT_ID=your-client-id
CLIENT_SECRET=your-client-secret
MCP_PORT=8000

Credentials: https://dashboard.groupdocs.cloud/#/applications

3. Run the server

Linux / macOS: ./run_mcp.sh
Windows (PowerShell): .\run_mcp.ps1
Windows (CMD): run_mcp.bat

Server URL: http://localhost:8000/mcp

Docker

docker build -t groupdocs-viewer-mcp .
docker run --rm -p 8000:8000 \
  -e CLIENT_ID=your-client-id \
  -e CLIENT_SECRET=your-client-secret \
  -e MCP_PORT=8000 \
  groupdocs-viewer-mcp

---

Integration examples

@modelcontextprotocol/inspector

1. npx @modelcontextprotocol/inspector
2. Transport: streamable HTTP, URL: http://127.0.0.1:8000/mcp

VS Code .vscode/mcp.json

{
  "servers": {
    "groupdocs-viewer-mcp-local": {
      "type": "http",
      "url": "http://127.0.0.1:8000/mcp"
    }
  }
}

Cursor

{
  "mcpServers": {
    "groupdocs-viewer-mcp-local": {
      "url": "http://127.0.0.1:8000/mcp"
    }
  }
}

Example: upload a PDF with file_upload_local, call viewer_get_info, then viewer_create_view with view_format HTML.

---

Reinitialize the virtual environment

Use init_mcp.sh / init_mcp.ps1 / init_mcp.bat, then start with run_mcp.*.

---

License

MIT License.