Groupdocs Total Mcp

io.github.groupdocs-totalPublished by GitHub account groupdocs-totalv26.7.3Updated Nov 3, 2026

MCP server for GroupDocs.Total — annotate, sign, convert, watermark, parse, redact documents via AI.

Overview

AI-generated overview

Local Docker-based MCP server exposing GroupDocs document tools so an assistant can annotate, sign, convert, redact, watermark, parse and compare files.

What you can do
  • Inspect documents: get file type, page count and size, and render pages as PNG previews
  • Annotate: add, list, update, remove annotations and replies, import and export them
  • Sign and verify: text, QR code, barcode, image and digital certificate signatures, plus signature search
  • Convert and transform: convert between 50+ formats, convert to or compose from Markdown, compare two documents, merge or split files
  • Parse and extract: text, images, tables, metadata and barcodes
  • Redact, watermark and metadata: regex text redaction, image-area blackout, annotation redaction, add/search/remove watermarks, read or strip metadata
When to use it
Useful when an assistant needs to work with real document files on your machine — preparing contracts for signature, converting Office files to PDF, extracting tables or text, cleaning metadata, or applying watermarks and redactions in bulk.
Requirements
Runs as a local Docker container pulled from GHCR (multi-arch); Docker must be installed and a host folder mounted for input and output files. Set GROUPDOCS_MCP_STORAGE_PATH for the base folder, optionally GROUPDOCS_MCP_OUTPUT_PATH, and GROUPDOCS_LICENSE_PATH pointing to a mounted GroupDocs license file. Without a license it runs in evaluation mode.
Before you install
The server writes, modifies and deletes files in the mounted folder, including redaction, metadata stripping and watermark removal, so point it at a dedicated directory. A license file path is requested via GROUPDOCS_LICENSE_PATH; without one, evaluation limits apply (output watermarks, page limits, per-process document-open caps). Signing with a digital certificate requires supplying that certificate.

Installation

In SourceWeft

  1. Open Groupdocs Total Mcp in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

GroupDocs.Total MCP Server

Unified MCP server that exposes the full GroupDocs.Total document-processing toolset — annotation, comparison, conversion, markdown, merging, metadata, parsing, redaction, signing, watermarking, plus cross-product document inspection and page preview — as AI-callable tools for Claude, Cursor, GitHub Copilot, and other MCP agents. Designed to be used standalone — attaching the individual product MCPs (GroupDocs.Annotation.Mcp, GroupDocs.Signature.Mcp, …) alongside is unnecessary and would duplicate tools.

[GHCR] [MCP Registry] [Integration tests] [License: MIT]

Installation

[Install in VS Code] [Install in VS Code Insiders] [Add to Cursor] [Codex CLI setup] [Windsurf setup]

One-click installs pre-fill every supported setting: edit the placeholder host path in the volume mount after install; an empty GROUPDOCS_LICENSE_PATH runs in evaluation mode.

More clients - ready-made configs for Claude Code, Codex CLI, Visual Studio 2022, Cursor, Windsurf, Cline, and JetBrains Rider live in install/generated/.

This product ships via Docker (GHCR) only. GroupDocs.Total bundles every engine (including Parser's embedded ONNX models), so the packed tool far exceeds NuGet.org's 250 MB limit - so dnx and dotnet tool install are not available for it. When upstream ships a slimmer engine that packs under the limit, the standard NuGet flow returns (see the revert notes in .github/workflows/publish_prod.yml).

bash
docker run --rm -i \  -v $(pwd)/documents:/data \  ghcr.io/groupdocs-total/total-net-mcp:latest

Images are multi-arch (linux/amd64 + linux/arm64), tagged latest plus an immutable version tag per release (e.g. :26.9.0). To pin, replace :latest with the version tag - recommended for shared configs and CI.

Available MCP Tools (38 total)

Tools follow {Product}{Verb}{Noun} PascalCase to disambiguate across domains. Cross-product tools stay short.

Cross-product (2)

ToolDescription
GetDocumentInfoFile type, page count, size as JSON for any document (170+ formats via Parser engine)
GetDocumentPageImageRender up to 5 pages as inline PNG images (Viewer engine, 170+ formats)

Annotation (8)

ToolDescription
AnnotationAddAnnotationAdd textfield / area / point / arrow / highlight / underline / strikeout annotation
AnnotationGetAnnotationsList all annotations as JSON (id, type, message, page, box, user, replies)
AnnotationUpdateAnnotationModify an existing annotation's message and/or bounding box by id
AnnotationRemoveAnnotationsRemove annotations by id list, or all
AnnotationAddReplyAdd a reply / comment thread to an existing annotation
AnnotationRemoveRepliesRemove replies by id, by user name, or all
AnnotationImportAnnotationsImport annotations from XML or another annotated document
AnnotationExportAnnotationsExtract annotations to XML (re-importable via Import)

Signature (7)

ToolDescription
SignatureSignSign with text / qrcode / barcode / digital certificate signature
SignatureVerifyVerify signatures, return JSON validity report
SignatureSearchTextSignaturesFind embedded text signatures with optional substring filter
SignatureSearchBarcodesFind barcode signatures with optional decoded-text filter
SignatureSearchQrCodesFind QR code signatures with optional decoded-text filter
SignatureSearchDigitalSignaturesFind digital certificate signatures (signer, issuer, validity)
SignatureSearchImageSignaturesFind embedded image signatures, return as base64 PNGs

Comparison (1) · Conversion (2) · Markdown (2)

ToolDescription
ComparisonCompareCompare two documents, produce annotated diff document
ConversionConvertConvert between PDF, Office, image, HTML, and 50+ more formats
ConversionGetSupportedFormatsList every output format the document can be converted TO
MarkdownConvertToMarkdownConvert PDF / Office / EPUB / MOBI to clean Markdown
MarkdownComposeFromMarkdownCompose DOCX / PDF / HTML from Markdown source

Merger (2) · Metadata (2)

ToolDescription
MergerMergeCombine up to 4 documents into one (PDF / Office / HTML / images)
MergerSplitSplit a document into separate files by page numbers
MetadataReadMetadataRead all metadata properties (author, title, EXIF, XMP, IPTC, custom) as JSON
MetadataRemoveMetadataStrip all metadata and save a clean copy

Parser (5) · Redaction (4) · Watermark (3)

ToolDescription
ParserExtractTextExtract plain text (page-by-page or whole document)
ParserExtractImagesExtract embedded images and save them to storage
ParserExtractTablesExtract tables as Markdown or JSON
ParserExtractMetadataExtract metadata fields (separate from MetadataReadMetadata; Parser-format-aware)
ParserExtractBarcodesFind and decode embedded barcodes in document pages
RedactionRedactTextRedact text matching a regex with a replacement string
RedactionRedactImageAreaBlack out a rectangular region of a document page
RedactionRedactAnnotationsRedact or delete annotations matching a regex
RedactionEraseMetadataStrip metadata via the Redaction engine (handles formats Metadata can't)
WatermarkAddWatermarkAdd a text watermark with configurable font, rotation, opacity
WatermarkSearchWatermarksList existing watermarks (type, text, position, page) as JSON
WatermarkRemoveWatermarksRemove watermarks by text filter or all

Example prompts

  • "Convert contract.docx to PDF, then sign it with a QR code containing 'Signed by Alice', then watermark every page with 'CONFIDENTIAL'"
  • "Compare old.pdf and new.pdf, list every annotation in the result, and preview page 1"
  • "Extract all tables from report.xlsx as Markdown and redact every email address"
  • "Read the metadata of every PDF in /uploads, then strip it, then preview the cleaned page 1 of each"
  • "Find every barcode in invoice.pdf, then merge invoice.pdf + receipt.pdf into combined.pdf"

Licensing

The MCP server itself is MIT; the underlying GroupDocs engines require a license for production use. Without one the server runs in evaluation mode:

  • Each bundled engine applies its own evaluation limits: output watermarks / evaluation notices, page limits, and per-process document-open caps (1-15 documents depending on the engine).

To lift the limits, mount your GroupDocs.Total.lic into the container and point GROUPDOCS_LICENSE_PATH at it (see the Claude Desktop example above):

Configuration

VariableDescriptionDefault
GROUPDOCS_MCP_STORAGE_PATHBase folder for input and output filescurrent directory
GROUPDOCS_MCP_OUTPUT_PATH(Optional) separate folder for output filesGROUPDOCS_MCP_STORAGE_PATH
GROUPDOCS_LICENSE_PATHPath to GroupDocs license file(evaluation mode)

Usage with Claude Desktop

json
{  "mcpServers": {    "groupdocs-total": {      "command": "docker",      "args": ["run", "--rm", "-i", "-v", "/path/to/documents:/data", "ghcr.io/groupdocs-total/total-net-mcp:latest"]    }  }}

To use a license, add "-v", "/path/to/license-folder:/license", "-e", "GROUPDOCS_LICENSE_PATH=/license/GroupDocs.Total.lic" before the image name. To pin a version, replace :latest with the release tag.

Usage with VS Code / GitHub Copilot

Use the Install in VS Code button above, or add to .vscode/mcp.json (also in install/generated/vscode-mcp.json):

json
{  "inputs": [    {      "type": "promptString",      "id": "storage_path",      "description": "Base folder for input and output files.",      "password": false    }  ],  "servers": {    "groupdocs-total": {      "command": "docker",      "args": ["run", "--rm", "-i", "-v", "${input:storage_path}:/data", "ghcr.io/groupdocs-total/total-net-mcp:latest"]    }  }}

Usage with Docker Compose

bash
cd dockerdocker compose up

Edit docker/docker-compose.yml to point volumes at your local documents folder.

License

MIT — see LICENSE

Source: README.md at commit 96dde17

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

1
  1. v26.7.3LatestSep 16, 2026

Score

50% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts