PDF Reader MCP

Production-ready PDF processing server for AI agents with enterprise-grade capabilities. Extract text, images, and metadata with 5-10x faster parallel processing performance.

Features

Core Capabilities

• Text Extraction - Full document or specific pages with intelligent parsing
• Image Extraction - Base64-encoded images with complete metadata (width, height, format)
• Content Ordering - Y-coordinate based layout preservation for natural reading flow
• Metadata Extraction - Author, title, creation date, and custom properties
• Page Counting - Fast enumeration without loading full content
• Dual Sources - Local files (absolute or relative paths) and HTTP/HTTPS URLs
• Batch Processing - Multiple PDFs processed concurrently

Performance

• 5-10x faster than sequential processing with automatic parallelization
• 12,933 ops/sec error handling, 5,575 ops/sec text extraction
• Process 50-page PDFs in seconds with multi-core utilization
• Lightweight with minimal dependencies

Advanced Features

• Smart Pagination - Extract ranges like "1-5,10-15,20"
• Multi-Format Images - RGB, RGBA, Grayscale with automatic detection
• Path Flexibility - Windows, Unix, and relative paths all supported
• Error Resilience - Per-page error isolation with detailed messages
• Large File Support - Efficient streaming and memory management
• Type Safe - Full TypeScript with strict mode enabled

Usage

Basic Text Extraction

{
  "sources": [{
    "path": "documents/report.pdf"
  }],
  "include_full_text": true,
  "include_metadata": true,
  "include_page_count": true
}

Extract Specific Pages

{
  "sources": [{
    "path": "documents/manual.pdf",
    "pages": "1-5,10,15-20"
  }],
  "include_full_text": true
}

Absolute Paths

{
  "sources": [{
    "path": "C:\\Users\\John\\Documents\\report.pdf"
  }],
  "include_full_text": true
}

Extract Images with Natural Ordering

{
  "sources": [{
    "path": "presentation.pdf",
    "pages": [1, 2, 3]
  }],
  "include_images": true,
  "include_full_text": true
}

Images and text are returned in document order based on Y-coordinates, preserving natural reading flow.

Batch Processing

{
  "sources": [
    { "path": "C:\\Reports\\Q1.pdf", "pages": "1-10" },
    { "path": "/home/user/Q2.pdf", "pages": "1-10" },
    { "url": "https://example.com/Q3.pdf" }
  ],
  "include_full_text": true
}

All PDFs are processed in parallel automatically.

From URL

{
  "sources": [{
    "url": "https://arxiv.org/pdf/2301.00001.pdf"
  }],
  "include_full_text": true
}

Security & Sandboxing

By default the server can read any local file and fetch any HTTP(S) URL. Use these options to restrict access:

Restrict Filesystem Access

npx @sylphx/pdf-reader-mcp --allow-dir=/srv/pdfs --allow-dir=/data/reports

Or via environment variable:

MCP_PDF_ALLOWED_DIRS="/srv/pdfs:/data/reports" npx @sylphx/pdf-reader-mcp

Disable or Restrict HTTP

# Block all URL sources
npx @sylphx/pdf-reader-mcp --no-http

# Allowlist hosts
npx @sylphx/pdf-reader-mcp --allow-host=cdn.example.com --allow-host=files.internal

HTTP Transport (Remote Access)

Run as an HTTP server for remote access:

MCP_TRANSPORT=http npx @sylphx/pdf-reader-mcp

Performance Benchmarks

Parallel Processing Speedup

• 10-page PDF: ~2s → ~0.3s (5-8x faster)
• 50-page PDF: ~10s → ~1s (10x faster)
• 100+ pages: Linear scaling with CPU cores

Quality

• ✅ 103 tests passing
• ✅ 94%+ test coverage
• ✅ 98%+ function coverage
• ✅ Strict TypeScript
• ✅ Production ready