MCP PDF turns scattered PDF content into structured, AI-assisted intelligence you can act on. It combines multi‑library processing, AI-driven analysis, and enterprise-grade security to extract text, tables, charts, and metadata—and to deliver ready‑to‑use insights fast for business, research, and compliance workflows.

How to use

You work with MCP PDF through an MCP client or orchestration layer. Start by launching the local MCP PDF server or connecting to a remote MCP endpoint, then issue high‑level objectives like classify_content, summarize_content, and extract_tables. The system automatically selects the best extraction path, applies AI analysis, and returns a structured, AI‑ready output you can feed into dashboards, reports, or data stores. You can fetch content from web URLs, analyze documents across pages, and combine results from multiple tools to build comprehensive insights.

Typical usage patterns include: processing a financial report to detect health, extracting all tables with token overflow protection, summarizing complex documents, and performing a security assessment on sensitive files. For web sources, the system downloads and caches content to speed subsequent requests. You can also integrate with Claude Desktop or other editors for seamless, cross‑format document intelligence.

How to install

Prerequisites: you need a supported runtime and the required system tools installed. The following steps install MCP PDF locally and verify that it runs.

# 1️⃣ Clone repository
git clone https://github.com/rsp2k/mcp-pdf
cd mcp-pdf

# 2️⃣ Install with uv (fastest)
uv sync

# 3️⃣ Install system dependencies (Ubuntu/Debian)
sudo apt-get install tesseract-ocr tesseract-ocr-eng poppler-utils ghostscript

# 4️⃣ Verify installation
uv run python examples/verify_installation.py

# 5️⃣ Run the MCP server
uv run mcp-pdf

Optional: for Claude Desktop integration, install production or development components and add the MCP server to your Claude configuration.

# Production Installation (PyPI) example entry for Claude Desktop
# For personal use across all projects
claude mcp add -s local pdf-tools uvx mcp-pdf

# Development Installation (Source)
claude mcp add -s project pdf-tools-dev uv -- --directory /path/to/mcp-pdf run mcp-pdf

Additional configuration and notes

Claude Desktop integration example shows how to wire two MCP servers together: a local pdf-tools server running via uv and a project‑scoped pdf-tools‑dev server. To use these, include the following in your Claude Desktop config under mcpServers.

{
  "mcpServers": {
    "pdf-tools": {
      "command": "uv",
      "args": ["run", "mcp-pdf"],
      "cwd": "/path/to/mcp-pdf"
    },
    "office-tools": {
      "command": "mcp-office-tools"
    }
  }
}

Security and enterprise readiness

MCP PDF provides local processing, memory security, strict HTTPS validation, configurable access controls, audit logging, GDPR compliance, and SOC2 readiness. These features ensure that documents remain in your environment, sensitive data is cleaned, and processing is auditable and compliant.

Installation & enterprise setup

Use these quick start steps to get MCP PDF up and running in an enterprise environment. Docker and Claude Desktop integrations are supported for scalable deployments.

# Quick Start (Recommended)
# 1. Clone repository
git clone https://github.com/rsp2k/mcp-pdf
cd mcp-pdf

# 2. Install with uv (fastest)
uv sync

# 3. Install system dependencies (Ubuntu/Debian)
sudo apt-get install tesseract-ocr tesseract-ocr-eng poppler-utils ghostscript

# 4. Verify installation
uv run python examples/verify_installation.py

Docker Enterprise Setup

FROM python:3.11-slim
RUN apt-get update && apt-get install -y \
    tesseract-ocr tesseract-ocr-eng \
    poppler-utils ghostscript \
    default-jre-headless
COPY . /app
WORKDIR /app
RUN pip install -e .
CMD ["mcp-pdf"]

Claude Desktop Integration (config snippet)

{
  "mcpServers": {
    "pdf-tools": {
      "command": "uv",
      "args": ["run", "mcp-pdf"],
      "cwd": "/path/to/mcp-pdf"
    },
    "office-tools": {
      "command": "mcp-office-tools"
    }
  }
}

Development Environment

# Clone and setup
git clone https://github.com/rsp2k/mcp-pdf
cd mcp-pdf
uv sync --dev

# Quality checks
uv run pytest --cov=mcp_pdf_tools
uv run black src/ tests/ examples/
uv run ruff check src/ tests/ examples/
uv run mypy src/

# Run all 23 tools demo
uv run python examples/verify_installation.py

Tool showcase and capabilities

The platform includes a wide arsenal of specialized tools for document intelligence, content extraction, layout analysis, and manipulation. You can perform AI‑driven classification, summarization, health and security analyses, text and table extraction, OCR, image extraction, layout detection, chart extraction, watermark detection, and more.

# Example: run a few core tools in sequence
pdf_analysis = await classify_content("report.pdf")
summary = await summarize_content("report.pdf", summary_length="medium")
text = await extract_text("report.pdf")

Unified integration and ecosystem

MCP PDF is designed to work alongside MCP Office Tools for cross‑format processing. You can route text, tables, images, and structure between tools to build end‑to‑end workflows that handle PDFs and other formats with a single, unified intelligence API.

You can connect MCP PDF to various clients, including Claude Desktop, Jupyter notebooks, Python applications, RESTful services, and cloud workflows, to enable scalable document intelligence across teams and systems.

Notes and troubleshooting

If you encounter issues starting the server, verify system dependencies, ensure the correct working directory for local tools, and confirm that the start command matches the runtime environment you configured (uv, npx, or Python entry points). Check logs for practical hints about missing dependencies or permission problems and address them before retrying.

Available tools

classify_content

AI-powered document type classification and analysis

summarize_content

Intelligent summarization with key insights

analyze_pdf_health

Comprehensive quality assessment of a PDF

analyze_pdf_security

Security feature analysis and vulnerability detection

compare_pdfs

Advanced document comparison including text, structure, and metadata

extract_text

Multi-method text extraction with auto-chunking

extract_tables

Smart table extraction with token overflow protection

ocr_pdf

OCR for scanned documents with preprocessing options

extract_images

Image extraction and processing from PDFs

pdf_to_markdown

Structure-preserving PDF to Markdown conversion

analyze_layout

Page layout analysis and column detection

extract_charts

Chart, diagram, and visual element extraction

detect_watermarks

Watermark detection and analysis

is_scanned_pdf

Smart detection of scanned vs. text-based documents

get_document_structure

Document outline and structural analysis

extract_metadata

Metadata and statistics extraction

extract_form_data

Interactive PDF form data extraction

split_pdf

Intelligent document splitting at specified pages

merge_pdfs

Merging multiple PDFs with page range tracking

rotate_pages

Precise page rotation (90/180/270)