# PDF Reader MCP [Health: Active]

**Category:** 🔎 Search & Data Extraction  
**Repository:** https://github.com/hancengiz/read_pdf_as_text_mcp  
**GitHub Stars:** 1  
**npm Downloads (last month):** 1201  
**Views:** 0  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/fabriqa-ai-pdf-reader-mcp

## Description
MCP server for efficient PDF text extraction, search, and metadata retrieval for Claude Code

## Tools
Capabilities this server exposes over MCP:

- **read-pdf** — Extract text from a PDF file. Returns the full text content of the PDF with optional page filtering and text cleaning.
- **search-pdf** — Search for specific text within a PDF file. Returns matching text with context and page numbers.
- **pdf-metadata** — Extract metadata from a PDF file including title, author, page count, creation date, etc.

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "pdf-reader-mcp": {
    "command": "npx",
    "args": ["-y","@fabriqa.ai/pdf-reader-mcp@latest"]
  }
}
```

## Documentation

## What PDF Reader MCP MCP server does

PDF Reader MCP MCP server gives an MCP-compatible AI client tools for working with local PDF files. It supports three operations: extracting text, searching for terms, and reading document metadata. The server is intended for cases where loading a complete PDF into the model context would be wasteful, such as reviewing research papers, checking long reports, or locating a specific section.

The `read-pdf` tool accepts a file path and can return all pages or a selected range. It also supports optional text cleaning and optional metadata in the result. The `search-pdf` tool looks for a supplied query and accepts case-sensitive and whole-word matching options. The `pdf-metadata` tool returns document information such as title, author, page count, and dates when that information is present in the PDF.

## How it works

The server is implemented as a local Node.js process and communicates with the MCP client through stdio. It uses `pdf-parse` for text extraction and the Model Context Protocol SDK for tool communication. Each tool operates on a PDF path supplied in the request, so the client environment must be able to access the referenced file.

A practical workflow is to inspect metadata first, search for a relevant phrase, and then read only the pages needed for closer analysis. Page ranges can use forms such as `1-5`, `1,3,5`, or `all`. Search matching is case-insensitive and allows partial-word matches by default, unless the optional flags change that behavior.

## Setup and configuration

Install the published package globally with `npm install -g @fabriqa.ai/pdf-reader-mcp`, or clone the repository and run `npm install`. Node.js must be installed and available in the system PATH.

For Claude Code, the documented command is `claude mcp add pdf-reader npx @fabriqa.ai/pdf-reader-mcp@latest`. Manual configuration adds a `pdf-reader` entry to `~/.claude.json` with `npx` as the command and the package name as its argument. Claude Desktop uses the same command and argument in its platform-specific configuration file. An included script can update the Claude configuration automatically.

When troubleshooting, check that the PDF exists, prefer an absolute path, verify file permissions, and restart the client after changing its configuration.

## Tools and capabilities

- `read-pdf`: Extracts PDF text from all pages or a specified page range. Optional cleaning normalizes the extracted text, and metadata inclusion is enabled by default.
- `search-pdf`: Searches a PDF for a required query, with optional case-sensitive and whole-word matching.
- `pdf-metadata`: Retrieves available PDF metadata, including fields such as title, author, page count, and dates.

## Limitations and notes

Text extraction depends on what `pdf-parse` can process. Scanned PDFs that contain images rather than text may not produce useful extracted content, and complex layouts can reduce extraction quality.

The tools read files rather than modifying them. The client must provide a valid path to a locally accessible PDF. PDF Reader MCP MCP server is documented primarily for Claude Code, although its stdio transport can be configured in other MCP clients that support local servers.

_Full upstream README: https://allmcps.com/mcp/fabriqa-ai-pdf-reader-mcp/readme_