PDF Reader

v2.4.2File Systemsstable

Enables reading, searching, and metadata extraction from PDF files without loading the entire content into the context window. It provides efficient tools for text cleaning, page-specific extraction, and context-aware search results.

ai-agentai-toolsdocument-processingllm-toolmcp
Share:
723
Stars
0
Downloads
0
Weekly
0/5

What is PDF Reader?

PDF Reader is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to enables reading, searching, and metadata extraction from pdf files without loading the entire content into the context window. it provides efficient tools for text cleaning, page-specific extraction, ...

Enables reading, searching, and metadata extraction from PDF files without loading the entire content into the context window. It provides efficient tools for text cleaning, page-specific extraction, and context-aware search results.

This server falls under the File Systems and Search & Data Extraction categories on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • Enables reading, searching, and metadata extraction from PDF

Use Cases

Read and search PDF files without loading entire documents into context. Extract metadata and page-specific content from PDFs efficiently. Perform context-aware searches across PDF documents.
SylphxAI

Maintainer

LicenseMIT License
Languagetypescript
Versionv2.4.2
UpdatedMay 22, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

NPM

npx -y @sylphx/pdf-reader-mcp

Manual Installation

npx -y @sylphx/pdf-reader-mcp

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use PDF Reader

The PDF Reader MCP Server by SylphxAI enables AI assistants to read, search, and extract content from PDF files without loading entire documents into the context window, using parallel processing that is 5-10x faster than sequential extraction. It exposes a single flexible read_pdf tool that handles local files, HTTP/HTTPS URLs, specific page ranges, metadata extraction, and base64-encoded embedded image extraction in one call. Document-heavy workflows — legal review, research paper analysis, financial report parsing — benefit from its page-specific access patterns that keep context usage minimal.

Prerequisites

  • Node.js 18+ installed (for npx execution)
  • An MCP client such as Claude Desktop, Claude Code, VS Code, or Cursor
  • PDF files accessible on the local filesystem or via HTTP/HTTPS URLs
  • Optional: MCP_PDF_ALLOWED_DIRS set to restrict filesystem access to specific directories
1

Add to Claude Code

Register the PDF reader MCP server with Claude Code using the CLI. This is the fastest way to get started.

claude mcp add pdf-reader -- npx @sylphx/pdf-reader-mcp
2

Configure Claude Desktop

Add the server to your Claude Desktop configuration file. No API keys are required — the server reads local files and public URLs.

{
  "mcpServers": {
    "pdf-reader": {
      "command": "npx",
      "args": ["-y", "@sylphx/pdf-reader-mcp"]
    }
  }
}
3

Restrict file system access (optional but recommended)

Set MCP_PDF_ALLOWED_DIRS to a colon-separated list of directories the server is allowed to read from. This prevents accidental access to sensitive files.

{
  "mcpServers": {
    "pdf-reader": {
      "command": "npx",
      "args": ["-y", "@sylphx/pdf-reader-mcp"],
      "env": {
        "MCP_PDF_ALLOWED_DIRS": "/Users/you/Documents:/Users/you/Downloads"
      }
    }
  }
}
4

Disable HTTP access for offline/security use

If you want to prevent the server from fetching PDFs from external URLs, pass the --no-http flag.

{
  "mcpServers": {
    "pdf-reader": {
      "command": "npx",
      "args": ["-y", "@sylphx/pdf-reader-mcp", "--no-http"]
    }
  }
}
5

Verify the setup

Ask Claude to read the metadata from a known PDF file to confirm the server is responding correctly.

PDF Reader Examples

Client configuration

Claude Desktop configuration for the PDF Reader MCP server with directory access restrictions.

{
  "mcpServers": {
    "pdf-reader": {
      "command": "npx",
      "args": ["-y", "@sylphx/pdf-reader-mcp"],
      "env": {
        "MCP_PDF_ALLOWED_DIRS": "/Users/you/Documents:/Users/you/Downloads",
        "MCP_PDF_ALLOW_HTTP": "true"
      }
    }
  }
}

Prompts to try

Example prompts for extracting content from PDFs using the pdf-reader server.

- "Read pages 1-5 of /Users/me/Documents/contract.pdf and summarize the key clauses"
- "Extract the metadata (author, title, creation date) from report.pdf"
- "Search for mentions of 'indemnification' in legal_agreement.pdf"
- "Fetch and read https://arxiv.org/pdf/2310.01234 and explain the main contributions"
- "Compare the executive summaries on page 1 of Q1.pdf and Q2.pdf"

Troubleshooting PDF Reader

Error: file path not allowed

This occurs when MCP_PDF_ALLOWED_DIRS is set and the requested file is outside the allowed paths. Add the file's parent directory to MCP_PDF_ALLOWED_DIRS, or remove the env var to allow unrestricted access.

HTTP PDF fetch fails with network error

Verify MCP_PDF_ALLOW_HTTP is not set to false and the URL is publicly accessible. For restricted hosts, add the domain to MCP_PDF_ALLOWED_HOSTS. Corporate proxies may block direct HTTP requests.

Large PDF returns incomplete text

Use the pages parameter to request specific page ranges (e.g., '1-10,25-30') rather than full document extraction. This reduces memory usage and avoids response size limits.

Frequently Asked Questions about PDF Reader

What is PDF Reader?

PDF Reader is a Model Context Protocol (MCP) server that enables reading, searching, and metadata extraction from pdf files without loading the entire content into the context window. it provides efficient tools for text cleaning, page-specific extraction, and context-aware search results. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install PDF Reader?

Install via npm with the command: npx -y @sylphx/pdf-reader-mcp. Then add the server configuration to your AI client's JSON config file (e.g., claude_desktop_config.json or .cursor/mcp.json).

Which AI clients work with PDF Reader?

PDF Reader works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is PDF Reader free to use?

Yes, PDF Reader is open source and available under the MIT License license. You can use it freely in both personal and commercial projects.

Browse More File Systems MCP Servers

Explore all file systems servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "pdf-reader-mcp-server": { "command": "npx", "args": ["-y", "@sylphx/pdf-reader-mcp"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use PDF Reader?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides