Log inSubmit server
Nutrient Document Engine MCP Server logo

Nutrient Document Engine MCP Server

by PSPDFKit

Self-hosted MCP access to Nutrient Document Engine for document extraction, editing, redaction, annotations, and discovery.

InstallRepositoryWebsite
Full README Compare alternatives Embed widget Claim this listing

Install Nutrient Document Engine MCP Server

Pick your client, then add it in one click or paste the config.

STDIONode.js
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "pspdfkit-nutrient-document-engine-mcp-server": {
      "command": "npx",
      "args": [
        "-y",
        "@nutrient-sdk/document-engine-mcp-server"
      ],
      "env": {
        "DASHBOARD_USERNAME": "YOUR_VALUE_HERE",
        "DASHBOARD_PASSWORD": "YOUR_VALUE_HERE",
        "DOCUMENT_ENGINE_BASE_URL": "YOUR_VALUE_HERE",
        "DOCUMENT_ENGINE_API_AUTH_TOKEN": "YOUR_VALUE_HERE",
        "MCP_TRANSPORT": "YOUR_VALUE_HERE",
        "MCP_HOST": "YOUR_VALUE_HERE",
        "MCP_HTTP_AUTH_TOKEN": "YOUR_VALUE_HERE",
        "ACTIVATION_KEY": "YOUR_VALUE_HERE"
      }
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

Values are written into the snippet above so you can paste it as is. They never leave your browser.

We couldn’t confirm it starts correctly

We ran the install command but it didn't respond within our test window — this can mean a slow first-time install rather than a real problem.

npx -y @nutrient-sdk/document-engine-mcp-server

No response to initialize.

This is an experimental automated check and can have false negatives — missing environment variables, a slow cold install, etc. It doesn’t necessarily mean something’s wrong. Last checked 16d ago.

Overview

PSPDFKit/nutrient-document-engine-mcp-server connects MCP-compatible AI agents to a self-hosted Nutrient Document Engine instance. It exposes tools for extracting content, searching and rendering documents, editing PDFs, handling forms, applying redactions, managing annotations, and inspecting document metadata. The server runs over stdio by default and also supports optional HTTP transport with bearer-token protection for network access. Choose it when document data must remain in an on-premises or private-cloud deployment rather than using Nutrient's hosted document workflow service.

Use cases

  • Extract text and structured data from uploaded documents
  • Redact sensitive values with preset patterns or custom regular expressions
  • Split, watermark, duplicate, and fill PDF documents
  • Search documents and inspect metadata through an AI client
  • Manage annotations and render document pages as images

Key features

  • Self-hosted Nutrient Document Engine connectivity
  • Content extraction, search, rendering, and form data access
  • PDF splitting, duplication, watermarking, and form filling
  • Pattern-based and regex-based redaction
  • Annotation management and document discovery
  • Stdio and optional authenticated HTTP transport

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked 1d ago

Recent health check succeeded.

20/21 checks healthy over the last 78d

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 11d ago via OSV.dev · @nutrient-sdk/document-engine-mcp-server (npm)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

GitHub stars62Stargazers on the source repository.
npm downloads55Package downloads in the last 30 days.
Last commit2mo agoMost recent push to the default branch.
Install checkInconclusiveDidn't respond in our test window — often a slow first install.
Directory activity3 viewsConfig copies, upvotes, and views on AllMCPs.

How Nutrient Document Engine MCP Server works

What PSPDFKit/nutrient-document-engine-mcp-server does

PSPDFKit/nutrient-document-engine-mcp-server provides an MCP interface to Nutrient Document Engine, allowing an AI client to invoke document-processing operations through natural-language requests or direct tool calls. Documents are managed by the connected Document Engine instance, while the MCP server exposes the available operations to an MCP-compatible client.

The project is intended for self-hosted or private-cloud deployments where the operator controls the Document Engine location and document data residency. It is separate from Nutrient's cloud-oriented DWS MCP Server, which is designed for hosted API workflows.

How it works

The default connection method is stdio. An MCP client starts the server with npx, and the server uses configuration values to reach the Document Engine API. Documents are uploaded through the Document Engine dashboard at the configured local address, then made available for processing through the connected AI client.

An optional HTTP transport can expose the MCP endpoint. It binds to 127.0.0.1 by default. If configured with a non-loopback host, it requires a nonempty MCP_HTTP_AUTH_TOKEN, and clients must send that value as a bearer token on every /mcp request. A token can also be required for loopback HTTP deployments. The /health and /dashboard paths are not protected by this MCP bearer token; dashboard access uses its own username and password settings.

Setup and configuration

Running the PSPDFKit/nutrient-document-engine-mcp-server setup requires Node.js 18 or newer for the npx-based MCP process and Docker Compose to run Nutrient Document Engine locally. The documented Claude Desktop setup starts Document Engine with Docker Compose, then adds an MCP entry whose command is npx and whose package is @nutrient-sdk/document-engine-mcp-server.

The example configuration supplies DASHBOARD_USERNAME, DASHBOARD_PASSWORD, DOCUMENT_ENGINE_BASE_URL, and DOCUMENT_ENGINE_API_AUTH_TOKEN. After restarting the client, documents can be uploaded through the Document Engine dashboard and queried from the MCP conversation.

Document Engine uses an evaluation license by default. Removing evaluation limitations requires purchasing a Nutrient license and setting ACTIVATION_KEY in the Document Engine or Docker Compose configuration, then restarting Document Engine. The activation key is a Document Engine setting rather than an MCP client credential.

Tools and capabilities

The PSPDFKit/nutrient-document-engine-mcp-server README groups its document operations into these areas:

  • Content access: extract text, search documents, render pages as images, and extract form data or table structures.
  • Document editing: split and duplicate documents, add watermarks, and fill PDF forms.
  • Security: create and apply redactions using preset sensitive-data patterns, such as social security numbers and credit cards, or custom regular expressions.
  • Annotations: add, read, and delete highlights, notes, stamps, and other document annotations.
  • System operations: discover documents, retrieve metadata, and monitor health.

The repository also includes examples for LangGraph, OpenAI Agents SDK workflows, a document-chat interface, and a procurement agent. These examples demonstrate possible integrations; they are not required for the MCP server's basic operation.

Limitations and notes

It does not provide client-specific setup instructions for every platform.

HTTP authentication protects MCP requests only. Dashboard credentials remain separately configured, and /health is not covered by the MCP bearer token. Evaluation mode may impose watermarks or other limitations until a licensed Document Engine configuration is supplied.

Read the full README View source on GitHub

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about Nutrient Document Engine MCP Server

PSPDFKit/nutrient-document-engine-mcp-server is an MCP server that connects AI clients to a self-hosted Nutrient Document Engine instance. Its tools support document discovery, text extraction, search, rendering, PDF editing, form filling, redaction, annotation management, metadata retrieval, and health monitoring through stdio or optional HTTP transport.

Run Nutrient Document Engine with Docker Compose, then configure your MCP client to start `@nutrient-sdk/document-engine-mcp-server` with `npx`. The documented setup requires Node.js 18+, Docker Compose, the Document Engine URL and API token, and dashboard credentials.

Cursor setup should therefore follow its MCP configuration process using the server's stdio or HTTP transport, subject to Cursor's own MCP support.

Set `MCP_TRANSPORT=http` and provide `MCP_HTTP_AUTH_TOKEN` when binding to a non-loopback `MCP_HOST`. Clients must send `Authorization: Bearer <token>` on every `/mcp` request. Dashboard access instead uses `DASHBOARD_USERNAME` and `DASHBOARD_PASSWORD`.

Related MCP servers

All in Other Tools and Integrations All alternatives
Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewNutrient Document Engine MCP Server AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/pspdfkit-nutrient-document-engine-mcp-server?style=directory)](https://allmcps.com/mcp/pspdfkit-nutrient-document-engine-mcp-server)
HTML Embed
<a href="https://allmcps.com/mcp/pspdfkit-nutrient-document-engine-mcp-server"><img src="https://allmcps.com/api/badge/pspdfkit-nutrient-document-engine-mcp-server?style=directory" alt="Nutrient Document Engine MCP Server on AllMCPs" /></a>