Pdf Reader Mcp Server
@shuji-bonji
An MCP (Model Context Protocol) server specialized in deciphering PDF internal structures.
Overview
What is Pdf Reader Mcp Server?
Pdf Reader Mcp Server is an MCP (Model Context Protocol) server specialized in deciphering PDF internal structures. It provides 15 tools organized into three tiers—basic operations, structure inspection, and validation & analysis—focusing on reading and analyzing the internal structure of PDF documents rather than simple text extraction. It is designed for developers and quality assurance professionals working with PDF content, and it pairs with pdf-spec-mcp for specification-aware validation and analysis.
How to use Pdf Reader Mcp Server?
Install the server via npx @shuji-bonji/pdf-reader-mcp. For Claude Desktop, add it to claude_desktop_config.json as an MCP server. For Claude Code, run claude mcp add pdf-reader-mcp -- npx -y @shuji-bonji/pdf-reader-mcp. It can also be run from source by cloning the repository and running npm run build. Tools are invoked by calling their name with parameters such as file_path, query, or pages.
Key features of Pdf Reader Mcp Server
- 15 tools across three functional tiers
- Text extraction with Y-coordinate reading order
- Full-text search with surrounding context
- Image extraction as base64 with metadata
- Fetch and process remote PDFs via URL
- Inspect structure tree, tags, fonts, annotations, signatures
- Validate PDF/UA tag structure (8 checks)
- Validate metadata conformance (10 checks)
- Compare structural differences between two PDFs
Use cases of Pdf Reader Mcp Server
- Analyze internal PDF object structure and catalog dictionary
- Validate tagged PDF/UA accessibility compliance
- Check metadata conformance against PDF/A or other standards
- Compare structural changes between document versions
- Summarize PDF metadata, text, and image count for quick overview
FAQ from Pdf Reader Mcp Server
How does this server differ from typical PDF MCP servers?
Typical PDF MCP servers are thin wrappers for text extraction, while this server focuses on reading and analyzing the internal PDF structure. It can be paired with pdf-spec-mcp for specification-aware analysis and validation.
What dependencies does the server use?
It uses TypeScript with the MCP TypeScript SDK, pdfjs-dist for text/image extraction and tag tree parsing, pdf-lib for low-level object structure analysis, Vitest for testing, and Zod for input validation.
