PDF Reader MCP

Add to your client

Copy the config for your MCP client and paste it into its config file.

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

Step-by-step guides: Add to Claude Desktop · Add to Cursor · Add to Windsurf

Before you start

• Node.js 22.13.0 or higher (required by pdfjs-dist v6)
• npx (bundled with Node.js) to run @sylphx/pdf-reader-mcp without a global install
• No credentials or API key needed — auth is none
• Optional: a directory to confine reads to (passed via --allow-dir), and network access if reading PDFs from URLs

About PDF Reader MCP

Tools & capabilities (1)

When to use it

• Use it when an agent needs to read and summarize a local PDF report, contract, or invoice as structured text.
• Use it when you want to pull specific page ranges out of a large PDF instead of the whole document.
• Use it when you need to fetch and parse a PDF directly from a URL without downloading it yourself first.
• Use it when you need PDF metadata (title, author, page count) for cataloging or routing logic.
• Use it when you want to extract embedded images (as Base64 with dimensions) from a PDF for downstream processing.
• Use it when you need to confine PDF access to a specific working directory for safety in an agent loop.

Quick setup

1. 1Ensure Node.js 22.13.0+ is installed.
2. 2Add the server to your MCP client config with command `npx` and args `["@sylphx/pdf-reader-mcp"]`.
3. 3Optionally pass `--allow-dir=/path/to/pdfs` (repeatable) to confine filesystem reads, and `--allow-host=domain` or `--no-http` to control URL access.
4. 4Restart your MCP client (e.g. Claude Desktop) so it picks up the new server.
5. 5Verify by asking the agent to read a known local PDF and confirm it returns text or metadata.

PDF Reader MCP FAQ