- Docling MCP @docling-project/docling-mcpConvert PDFs and other documents to structured formats via Docling, for AI applications.3% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.768 GitHub stars.
- Google Surf @harimxchoi/google-surf-mcpWeb, academic and code search with graph RAG, data lineage, ontology and cross-project schema links.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.291 GitHub stars.
- reMarkable MCP Server @sammorrowdrums/remarkableRead, render, search, and manage reMarkable documents through MCP3% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.243 GitHub stars.
- MinerU Open MCP @opendatalab/mineru-open-mcpParse PDFs, images, doc, docx, ppt, pptx, xls, xlsx, html into Markdown using MinerU API.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.221 GitHub stars.
- Gemini @rlabs-inc/gemini-mcpGemini 3 MCP server with 30+ tools: images, video, research, TTS, code exec & CLI7% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.217 GitHub stars.
- Office Oxide @aimino-tech/office-oxide-mcpRust-native MCP server for Office document processing5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.163 GitHub stars.
- PDF @jztan/pdf-mcpAgentic RAG over one PDF or a whole folder: hybrid search, selective page reads, tables, OCR.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.149 GitHub stars.
- Rendercheck @parthibanrajasekaran/rendercheckCheck from a screenshot that text is really visible to users. Local, offline OCR.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.94 GitHub stars.
- Obsidian Vault MCP @luffysolution-svg/obsidian-vault-mcpLocal Zotero, MinerU, Obsidian literature workflow with transactional Analysis notes.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.93 GitHub stars.
- pdfmux @nameetp/pdfmuxPDF-to-Markdown extraction that audits its own output and flags any extractor's silent drops.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.82 GitHub stars.
- Nutrient DWS MCP Server @pspdfkit/nutrient-dws-mcp-serverNutrient Processor + Data Extraction MCP: convert, OCR, sign, redact, and extract PDFs/Office docs.3% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.71 GitHub stars.
- Nutrient Document Engine MCP Server @pspdfkit/nutrient-document-engine-mcp-serverMCP server for self-hosted Nutrient Document Engine: extract, edit, redact, and annotate documents.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.62 GitHub stars.
- DocSlicer @docslicer/docslicerAnswer questions from documents too large to fit in context, reading only the sections you need.3% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from MCPB.35 GitHub stars.
- Documentation Extractor @nirholas/extract-llms-docsExtract llms.txt from any docs site - Mintlify Docusaurus GitBook parser3% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from a package registry.34 GitHub stars.
- Pdfco @mcparmory/pdfcoExtract data, edit, convert, and parse PDF documents with OCR and AI6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.32 GitHub stars.
- RAGStack @hatmanstack/ragstackSearch, chat, upload, and scrape a serverless RAGStack knowledge base on AWS.7% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.25 GitHub stars.
- MCP Document Reader @xt765/mcp-documents-readerAn MCP enabled multi-format document reader supporting DOCX, PDF, TXT, and Excel files4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.23 GitHub stars.
- reMarkable MCP Server @wavyrai/rm-mcpAccess your reMarkable tablet - read documents, browse files, extract text and OCR handwritten notes4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.21 GitHub stars.
- DeepSeek Visionary @xlight/deepseek-visionaryDeepSeek web vision model as an MCP server with browser auto-login.18% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from MCPB.18 GitHub stars.
- Mistral @swih/mistral-mcpMistral MCP server: extract invoices and documents from text, Markdown or OCR into typed JSON.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.15 GitHub stars.
- SharePoint MCP Server @ravikant1918/sharepoint-mcpMCP server for Microsoft SharePoint — manage folders, documents, and metadata.6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.12 GitHub stars.
Document data extraction MCP servers
A server here extracts text or fields from documents, receipts or scans (OCR).
Not here: converting file formats.
292 remote281 localTagged by Odel's AI, which can make mistakes
Most popular in Document data extraction
By GitHub stars, at most two per publisher- Number 1: anymd@sylphxai/anymdAny file → clean Markdown for AI agents: PDF, Office, EPUB, HTML, images, audio/video. Local MCP.3% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.1,035 GitHub stars.
- Number 2: Citra@sylphxai/citraCitra — PDF answers with page-level proof: structure, tables, OCR, and page citations.3% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.1,030 GitHub stars.
- Number 3: PDF Reader MCP@sylphxai-io/pdf-reader-mcpEvidence-first PDF MCP. Agent Document Twin with citeable page+bbox evidence.3% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.1,028 GitHub stars.
