Viewer for .NET Viewer for Java Viewer for Node.js Viewer for Python

Viewer for AI Agents (MCP)

Document page rendering for AI agents via MCP server

Give Claude, Cursor or Copilot eyes for your documents. The server renders any page of 170+ formats as an image the agent can read, including charts, diagrams and scans, and nothing is uploaded.

NuGet Download Start Free Trial

One-click install: VS Code Cursor All clients

Version 26.9.0 released

See what’s new

{
  "mcpServers": {
    "groupdocs-viewer": {
      "command": "dnx",
      "args": ["GroupDocs.Viewer.Mcp", "--yes"],
      "env": {
        "GROUPDOCS_MCP_STORAGE_PATH": "/path/to/documents"
      }
    }
  }
}

Then ask: Render page 7 of reportA.pdf and read the revenue chart

dnx GroupDocs.Viewer.Mcp --yes

GroupDocs.Viewer MCP server at a glance

An MCP server that gives AI agents the GroupDocs.Viewer engine as tools they can call

Let the agent see the page

Text extraction misses charts, diagrams, stamps and scanned pages. GroupDocs.Viewer renders the page faithfully as PNG, so a vision-capable model can read it the way a person would.

Local by design

Your AI client starts the server as a local process over stdio. It opens no ports, sends no telemetry, and reads and writes only the folder you configure, so contracts, records and internal documents never leave your machine.

Works with every MCP client

Use it from Claude Desktop, Claude Code, VS Code with GitHub Copilot, Visual Studio 2022, Cursor, Windsurf, Cline, JetBrains Rider and Codex CLI. One-click installs and ready-made configurations are available for each client.

Runs where your agent runs

GroupDocs.Viewer MCP server works on Windows, Linux and macOS, natively with .NET 10 or in Docker

One command with the .NET 10 SDK, no install step
Windows, Linux and macOS

NuGet download

Nothing to install besides Docker
linux/amd64 and linux/arm64, native dependencies bundled

Get the image

One-click install with a pre-filled configuration
Set your documents folder after install

Install in VS Code

One-click install with a pre-filled configuration
Runs in evaluation mode until you add a license

Add to Cursor

Supported file formats

GroupDocs.Viewer MCP server works with the following file formats.

Microsoft Office, OpenDocument and text formats

Word: DOC, DOCX, DOCM, DOT, DOTX, DOTM, RTF, TXT
Excel: XLS, XLSX, XLSM, XLSB, XLTM, XLT, XLTM, XLTX
PowerPoint: PPT, PPTX, PPS, PPSX, PPSM, POT, POTM, POTX, PPTM
Project: MPP, MPT, MPX
Outlook: MSG, EML, EMLX, PST, OST
OneNote: ONE Windows-supported target frameworks (net462 / net6.0-windows / etc.)
OpenDocument: ODT, OTT, ODS, ODP, OTP, OTS, ODG
Fixed Page Layout: PDF, TEX, XPS, OXPS
e-Books: EPUB, MOBI, DjVu
Delimiter-Separated Values: CSV, TSV

Images, Graphics & Diagrams

Raster images: BMP, GIF, JPG, PNG, TIFF, WebP, DNG, DIB, Jpeg2000 family
Windows Icon: ICO
Scalable Vector Graphics: SVG, AI, CDR, CMX, IGS, SVGZ Windows-supported target frameworks (net462 / net6.0-windows / etc.)
Adobe Photoshop: PSD, PSB Windows-supported target frameworks (net462 / net6.0-windows / etc.)
Stereo Lithography (3D Printing): STL
Medical Imaging: DICOM
Plotter Documents: PLT, HPG
Autodesk Design Web Formats: DWF, DWG
AutoCAD Drawing: DWT, IFC, STL, CF2

Other

Web: HTML, MHT, MHTML, XML
Metafile: WMF, EMF, CGM, EMZ, WMZ
Visio: VSD, VDX, VSS, VSSX, VSX, VST, VSTX, VTX, VSDX, VDW, VSTM, VSSM, VSDM
Project: MPP, MPT, MPX
PostScript: PS, EPS
Archives: ZIP, TAR, BZ2, GZ, RAR, RAR5
Other: VCF, VCARD, NUMBERS, NSF, OBJ
C/C++/C# Files: C, CC, C# , CPP, CXX, CS, H, HH, M, MM
Java/JavaScript Files: JAVA, JS, JSON, PROPERTIES

What the MCP server can do

Render any page of 170+ formats as an image your agent can read

Render pages as images

The render_page tool returns a page as PNG inline and saves a copy to storage.

Page count and sizes

The get_view_info tool returns file type, page count and page dimensions without rendering.

Charts, diagrams and scans

Ask questions about visual content that text extraction cannot reach.

170+ formats

Render PDF, DOCX, XLSX, PPTX, ODT, RTF, HTML, EML, MSG, CAD and more.

Protected documents

Pass the password and render pages of encrypted files.

Images for reports and tickets

Save page images to attach to tickets, reviews or reports.

Faithful rendering

Fonts, layout and embedded images are rendered by the GroupDocs engine, not approximated.

Pick the page first

Check the page count, then render only the pages the question needs.

Flexible licensing

Start in evaluation mode, then use a license file or metered keys. The get_license_status tool shows which one is active.

What you can ask your agent

Typical GroupDocs.Viewer MCP scenarios, each with the prompt that drives it

Read charts and scanned pages

Numbers in a chart are invisible to text extraction. The agent renders the page and reads the figures the way a person would. Step by step: guide and blog walkthrough.

Prompt

On page 7 of the annual report there is a revenue chart.
What were the figures for each quarter?

Let the agent check a document visually

Ask the agent to look before it answers. It checks the page count, renders the page you need and describes what it sees. Step by step: guide and blog walkthrough.

Prompt

How many pages does the report have? Then show me the last one.

Create page images for reports and tickets

Bug reports, reviews and audits need screenshots of specific pages. The agent renders them and tells you where the files are. Step by step: guide and blog walkthrough.

Prompt

Render pages 1, 4 and 9 of the specification
and tell me where the files are

Run on-premise with Docker

For regulated environments, run the server inside your perimeter. Mount the documents folder and the license, and pin the image version for change control. Architecture and security model: on-premise deployment.

docker-compose.yml

services:
  groupdocs-viewer:
    image: ghcr.io/groupdocs-viewer/viewer-net-mcp:26.9.0
    volumes:
      - "/srv/documents:/data"
      - "/srv/licenses:/license:ro"
    environment:
      GROUPDOCS_MCP_STORAGE_PATH: /data
      GROUPDOCS_LICENSE_PATH: /license/GroupDocs.Total.lic
    stdin_open: true

Install in your AI client

Copy the configuration for your client, or use the one-click links at the top of the page

Register the server

Every client starts the server the same way. Set GROUPDOCS_MCP_STORAGE_PATH to the folder with your documents, then restart the client. Leave GROUPDOCS_LICENSE_PATH empty for evaluation mode. Configurations for Visual Studio 2022, Windsurf, Cline, JetBrains Rider and Codex CLI are in the installation guide.

Claude Desktop Claude Code VS Code Cursor Docker

{
  "mcpServers": {
    "groupdocs-viewer": {
      "type": "stdio",
      "command": "dnx",
      "args": ["GroupDocs.Viewer.Mcp", "--yes"],
      "env": {
        "GROUPDOCS_MCP_STORAGE_PATH": "/path/to/documents",
        "GROUPDOCS_LICENSE_PATH": ""
      }
    }
  }
}

claude mcp add groupdocs-viewer \
  -e GROUPDOCS_MCP_STORAGE_PATH=/path/to/documents \
  -- dnx GroupDocs.Viewer.Mcp --yes

{
  "servers": {
    "groupdocs-viewer": {
      "type": "stdio",
      "command": "dnx",
      "args": ["GroupDocs.Viewer.Mcp", "--yes"],
      "env": {
        "GROUPDOCS_MCP_STORAGE_PATH": "/path/to/documents",
        "GROUPDOCS_LICENSE_PATH": ""
      }
    }
  }
}

{
  "mcpServers": {
    "groupdocs-viewer": {
      "command": "dnx",
      "args": ["GroupDocs.Viewer.Mcp", "--yes"],
      "env": {
        "GROUPDOCS_MCP_STORAGE_PATH": "/path/to/documents",
        "GROUPDOCS_LICENSE_PATH": ""
      }
    }
  }
}

docker run --rm -i \
  -v /path/to/documents:/data \
  ghcr.io/groupdocs-viewer/viewer-net-mcp:latest

Document page rendering for AI agents via MCP server

Show an AI agent a document page

GroupDocs.Viewer MCP server at a glance

Let the agent see the page

Local by design

Works with every MCP client

Runs where your agent runs

NuGet package

Docker image

VS Code and GitHub Copilot

Cursor

Supported file formats

Microsoft Office, OpenDocument and text formats

Images, Graphics & Diagrams

Other

What the MCP server can do

What you can ask your agent

Read charts and scanned pages

Prompt

Let the agent check a document visually

Prompt

Create page images for reports and tickets

Prompt

Run on-premise with Docker

docker-compose.yml

Install in your AI client

Register the server

Ready to get started?

Useful resources

Document page rendering for AI agents via MCP server

Show an AI agent a document page

GroupDocs.Viewer MCP server at a glance

Let the agent see the page

Local by design

Works with every MCP client

Runs where your agent runs

NuGet package

Docker image

VS Code and GitHub Copilot

Cursor

Supported file formats

Microsoft Office, OpenDocument and text formats

Images, Graphics & Diagrams

Other

What the MCP server can do

What you can ask your agent

Read charts and scanned pages

Prompt

Let the agent check a document visually

Prompt

Create page images for reports and tickets

Prompt

Run on-premise with Docker

docker-compose.yml

Install in your AI client

Register the server

Ready to get started?

Useful resources

Temporary license tips