PDF Reader MCP Server (@shtse8/pdf-reader-mcp)
@shtse8
📄 The PDF intelligence layer for AI agents — Agent Document Twin, evidence-first extraction, visual crops, OCR provenance, trust reports, and benchmark-gated releases. MCP server for Claude, Cursor, VS Code, and any MCP client.
Overview
What is PDF Reader MCP Server?
PDF Reader MCP Server transforms PDFs into an Agent Document Twin: a linked, source-backed representation that AI agents can inspect, search, verify, crop, OCR, enrich, cite, and read with confidence. It is built for agents that need evidence and provenance, not just extracted text.
How to use PDF Reader MCP Server?
Install and run via npx: npx @sylphx/pdf-reader-mcp. Node.js 22.13 or higher is required. For Claude Desktop, add a configuration entry with the same command. A Docker image is also available from GitHub Container Registry. The default package works without any OCR models, vision models, or cloud credentials.
Key features of PDF Reader MCP Server
- One smart tool (
read_pdf) that auto-inspects and reads PDFs - Multi-layered Agent Document Twin: lossless PDF, visual, semantic, evidence, and agent layers
- Optional OCR and visual region analysis via provider adapters
- Built-in trust and accessibility reports for safety
- Evidence-first extraction with page numbers, bounding boxes, and crop IDs
- Benchmark-gated releases with deterministic quality checks
Use cases of PDF Reader MCP Server
- Read and summarize PDFs with page-level citations
- Extract and verify tables, charts, formulas, and figures
- Search PDF content with result snippets and bounding boxes
- Handle scanned PDFs by routing to configured OCR providers
- Protect agents from hidden text, spoofing, and unsafe links
FAQ from PDF Reader MCP Server
What runtime does PDF Reader MCP Server require?
Node.js `>=22.13
