{
  "slug": "read-pdfs-and-documents",
  "label": "read PDFs and documents",
  "blurb": "Turn PDFs, scans and Office files into text, tables and structured data.",
  "keywords": [
    "pdf",
    "ocr",
    "document parsing",
    "documents"
  ],
  "categories": [
    "research",
    "data-analytics",
    "content-media"
  ],
  "access": [
    "api",
    "mcp"
  ],
  "count": 28,
  "official": 5,
  "url": "https://agentswitchboard.dev/tasks/read-pdfs-and-documents",
  "browseUrl": "https://agentswitchboard.dev/browse?task=read-pdfs-and-documents",
  "agentsUrl": "https://agentswitchboard.dev/api/agents?task=read-pdfs-and-documents",
  "agents": [
    {
      "name": "PaddleOCR",
      "slug": "paddleocr",
      "url": "https://agentswitchboard.dev/agents/paddleocr",
      "iconUrl": "https://agentswitchboard.dev/icons/paddleocr.png?v=607bffa6",
      "description": "Converts documents, receipts, and images into structured data with OCR, layout analysis, and table recognition, exposed to agents via a Python API, CLI, and MCP server.",
      "provider": "PaddlePaddle",
      "categories": [
        "content-media",
        "data-analytics"
      ],
      "accessMethods": [
        "api",
        "cli",
        "mcp"
      ],
      "verified": true,
      "official": true,
      "featured": true,
      "authType": "none",
      "why": "Text Recognition: Extracts printed and handwritten text from images and scanned PDFs across more than 80 languages."
    },
    {
      "name": "MinerU",
      "slug": "mineru",
      "url": "https://agentswitchboard.dev/agents/mineru",
      "iconUrl": "https://agentswitchboard.dev/icons/mineru.png?v=b6cc740a",
      "description": "Converts PDFs, Office documents, and scanned pages into LLM-ready Markdown or JSON, preserving tables, formulas, and reading order for agent pipelines.",
      "provider": "OpenDataLab",
      "categories": [
        "data-analytics",
        "research"
      ],
      "accessMethods": [
        "api",
        "cli"
      ],
      "verified": true,
      "official": false,
      "featured": false,
      "authType": "apiKey",
      "why": "Layout Parsing: Detects columns, headers, and figures to reconstruct correct reading order from complex multi-column documents."
    },
    {
      "name": "LiteParse",
      "slug": "liteparse",
      "url": "https://agentswitchboard.dev/agents/liteparse",
      "iconUrl": "https://agentswitchboard.dev/icons/liteparse.png?v=dfecfc3e",
      "description": "Parses PDFs, Office files, and images locally into markdown, JSON with bounding boxes, or plain text for agents and RAG pipelines, from a CLI or Rust, Node.js, Python, and WASM libraries.",
      "provider": "LlamaIndex",
      "categories": [
        "data-analytics",
        "code-devtools"
      ],
      "accessMethods": [
        "cli",
        "api"
      ],
      "verified": true,
      "official": false,
      "featured": false,
      "authType": "none",
      "why": "Selective OCR: Runs bundled Tesseract OCR only where native text is missing, or sends pages to an HTTP OCR server such as EasyOCR or PaddleOCR."
    },
    {
      "name": "OXR (SoMark)",
      "slug": "somark-oxr",
      "url": "https://agentswitchboard.dev/agents/somark-oxr",
      "iconUrl": "https://agentswitchboard.dev/icons/somark-oxr.png?v=1b796e88",
      "description": "Parses PDFs and document images into Markdown or JSON with layout coordinates, tables, formulas and handwriting, via a self-hosted or free API. For agents and document pipelines.",
      "provider": "SoMark",
      "categories": [
        "data-analytics",
        "research"
      ],
      "accessMethods": [
        "api",
        "cli"
      ],
      "verified": true,
      "official": false,
      "featured": false,
      "authType": "none",
      "why": "Document parsing: Turns PDFs and document images into Markdown or JSON output, with layout coordinates included."
    },
    {
      "name": "Chandra",
      "slug": "chandra-ocr",
      "url": "https://agentswitchboard.dev/agents/chandra-ocr",
      "iconUrl": "https://agentswitchboard.dev/icons/chandra-ocr.png?v=d17eb79e",
      "description": "Converts PDFs and images into Markdown, HTML or JSON with layout preserved, handling tables, forms, handwriting and 90+ languages, via a CLI with local or vLLM inference.",
      "provider": "Datalab",
      "categories": [
        "data-analytics",
        "infrastructure"
      ],
      "accessMethods": [
        "cli",
        "api"
      ],
      "verified": true,
      "official": false,
      "featured": false,
      "authType": "none",
      "why": "Layout-Preserving OCR: Converts images and PDFs into Markdown, HTML or JSON with detailed layout information, and extracts images and diagrams with…"
    },
    {
      "name": "Docling",
      "slug": "docling",
      "url": "https://agentswitchboard.dev/agents/docling",
      "iconUrl": "https://agentswitchboard.dev/icons/docling.png?v=ed47b089",
      "description": "Parses PDF, DOCX, PPTX, XLSX, HTML, audio, and image files into a unified DoclingDocument and exports Markdown, HTML, DocTags, or lossless JSON for gen-AI pipelines.",
      "provider": "IBM Research / LF AI & Data",
      "categories": [
        "content-media",
        "data-analytics",
        "research"
      ],
      "accessMethods": [
        "api",
        "cli",
        "mcp"
      ],
      "verified": true,
      "official": false,
      "featured": false,
      "authType": "none",
      "why": "PDF Layout Understanding: Resolves page layout, reading order, table structure, code blocks, and formulas in PDFs, with OCR for scanned pages and…"
    },
    {
      "name": "MarkItDown",
      "slug": "markitdown",
      "url": "https://agentswitchboard.dev/agents/markitdown",
      "iconUrl": "https://agentswitchboard.dev/icons/markitdown.png?v=b809713c",
      "description": "Converts PDF, Office, HTML, image, audio, and ZIP files into structured Markdown for LLM pipelines, preserving headings, lists, tables, and links.",
      "provider": "Microsoft",
      "categories": [
        "content-media",
        "data-analytics"
      ],
      "accessMethods": [
        "cli",
        "mcp"
      ],
      "verified": true,
      "official": true,
      "featured": false,
      "authType": "none",
      "why": "Office Document Conversion: Converts PDF, Word, PowerPoint, Excel, EPUB, and HTML files to Markdown while preserving headings, lists, tables, and…"
    },
    {
      "name": "AnyDoc",
      "slug": "anydoc",
      "url": "https://agentswitchboard.dev/agents/anydoc",
      "iconUrl": "https://agentswitchboard.dev/icons/anydoc.png?v=adf818e3",
      "description": "Converts Word, PowerPoint, Excel, OpenDocument, RTF, EPUB, CSV, and PDF files to GitHub-Flavored Markdown for LLMs, via a CLI or Rust, Node, Python, and WASM libraries.",
      "provider": "Firecrawl",
      "categories": [
        "infrastructure"
      ],
      "accessMethods": [
        "cli",
        "api"
      ],
      "verified": true,
      "official": false,
      "featured": false,
      "authType": "none",
      "why": "Hosted OCR Fallback: Sends scanned or image-only PDFs to Firecrawl Parse for OCR when opted in with an API key, returning the same Markdown."
    },
    {
      "name": "Markdownify MCP",
      "slug": "markdownify-mcp",
      "url": "https://agentswitchboard.dev/agents/markdownify-mcp",
      "iconUrl": "https://agentswitchboard.dev/icons/markdownify-mcp.png?v=46896277",
      "description": "MCP server that converts PDFs, images, audio files, and web pages into clean Markdown for efficient LLM ingestion and RAG pipeline construction.",
      "provider": "zcaceres",
      "categories": [
        "research"
      ],
      "accessMethods": [
        "mcp"
      ],
      "verified": true,
      "official": false,
      "featured": false,
      "authType": "none",
      "why": "PDF to Markdown: Convert PDF documents to clean Markdown preserving headings, tables, and text formatting"
    },
    {
      "name": "Presenton",
      "slug": "presenton",
      "url": "https://agentswitchboard.dev/agents/presenton",
      "iconUrl": "https://agentswitchboard.dev/icons/presenton.png?v=664175cf",
      "description": "Generates editable PPTX and PDF presentations from prompts, documents, or your own templates for teams and developers, via a self-hosted app, REST API, or MCP server.",
      "provider": "Presenton",
      "categories": [
        "content-media",
        "productivity"
      ],
      "accessMethods": [
        "api",
        "mcp"
      ],
      "verified": true,
      "official": false,
      "featured": false,
      "authType": "apiKey",
      "why": "PPTX and PDF Export: Exports fully editable PowerPoint (PPTX) and PDF decks."
    },
    {
      "name": "MangaTranslate",
      "slug": "mangatranslate",
      "url": "https://agentswitchboard.dev/agents/mangatranslate",
      "iconUrl": "https://agentswitchboard.dev/icons/mangatranslate.png?v=c533fc20",
      "description": "Translates manga pages and PDFs through a credit-metered REST API: OCR, text removal, background repair and typesetting, submitted as tasks and polled for translated image URLs.",
      "provider": "Ficory LLC",
      "categories": [
        "language",
        "content-media"
      ],
      "accessMethods": [
        "api"
      ],
      "verified": true,
      "official": false,
      "featured": false,
      "authType": "apiKey",
      "why": "Submit Translation: Submits a manga page or PDF to POST /api/v1/api/translate with a source and target language, returning a task for the translation…"
    },
    {
      "name": "Ragie",
      "slug": "ragie",
      "url": "https://agentswitchboard.dev/agents/ragie",
      "iconUrl": "https://agentswitchboard.dev/icons/ragie.png?v=1bc67b4c",
      "description": "Retrieval API for AI applications. Ingest PDFs, web pages, and files, then query with semantic search. Built-in chunking, embedding, and reranking pipeline.",
      "provider": "Ragie",
      "categories": [
        "research"
      ],
      "accessMethods": [
        "api"
      ],
      "verified": true,
      "official": false,
      "featured": false,
      "authType": "apiKey",
      "why": "Ingest PDFs, DOCX, web pages, and 20+ file types into a managed retrieval pipeline"
    },
    {
      "name": "Semantic Scholar MCP",
      "slug": "semantic-scholar-mcp",
      "url": "https://agentswitchboard.dev/agents/semantic-scholar-mcp",
      "iconUrl": "https://agentswitchboard.dev/icons/semantic-scholar-mcp.png?v=1d30b052",
      "description": "Searches millions of academic papers on Semantic Scholar and arXiv with citation analysis and full-text PDF extraction. No API key required.",
      "provider": "Community",
      "categories": [
        "research"
      ],
      "accessMethods": [
        "mcp"
      ],
      "verified": true,
      "official": false,
      "featured": false,
      "authType": "none",
      "why": "PDF Extraction: Downloads and extracts full text from arXiv and open-access PDFs for deep reading without leaving the agent context."
    },
    {
      "name": "WeKnora",
      "slug": "weknora",
      "url": "https://agentswitchboard.dev/agents/weknora",
      "iconUrl": "https://agentswitchboard.dev/icons/weknora.png?v=8859ab3f",
      "description": "Turns documents into a queryable knowledge base with RAG question answering, a ReAct agent orchestrating retrieval, MCP tools and web search, and an auto-generated wiki.",
      "provider": "Tencent",
      "categories": [
        "research",
        "customer-support",
        "communication"
      ],
      "accessMethods": [
        "api",
        "mcp",
        "cli",
        "browser-extension"
      ],
      "verified": true,
      "official": true,
      "featured": false,
      "authType": "apiKey",
      "why": "Document Ingestion: Parses PDF, Word, Excel, PPT, EPUB, images and more, with per-upload parser, chunking, OCR and VLM settings plus batch reparse."
    },
    {
      "name": "PageIndex",
      "slug": "pageindex",
      "url": "https://agentswitchboard.dev/agents/pageindex",
      "iconUrl": "https://agentswitchboard.dev/icons/pageindex.png?v=22343279",
      "description": "Indexes long documents as a reasoning-based hierarchical tree so agents retrieve by navigating structure instead of embedding similarity.",
      "provider": "Vectify AI",
      "categories": [
        "research",
        "vector-databases"
      ],
      "accessMethods": [
        "api",
        "mcp"
      ],
      "verified": true,
      "official": false,
      "featured": false,
      "authType": "apiKey",
      "why": "Tree Index Generation: Builds a hierarchical table-of-contents index over long documents that mirrors their actual section structure."
    },
    {
      "name": "RAGFlow",
      "slug": "ragflow",
      "url": "https://agentswitchboard.dev/agents/ragflow",
      "iconUrl": "https://agentswitchboard.dev/icons/ragflow.png?v=f608a034",
      "description": "Open-source RAG engine with deep document parsing (tables, charts, images) and an official MCP server. Deploy via Docker, call via REST API. 80k GitHub stars.",
      "provider": "InfiniFlow",
      "categories": [
        "research",
        "infrastructure"
      ],
      "accessMethods": [
        "api",
        "mcp"
      ],
      "verified": true,
      "official": false,
      "featured": false,
      "authType": "apiKey",
      "why": "Deep Document Parse: Parse complex documents — tables, charts, multi-column layouts — with layout-aware chunking for accurate retrieval."
    },
    {
      "name": "Unlayer MCP",
      "slug": "unlayer-mcp",
      "url": "https://agentswitchboard.dev/agents/unlayer-mcp",
      "iconUrl": "https://agentswitchboard.dev/icons/unlayer-mcp.png?v=091e45f5",
      "description": "Lets coding agents build email, landing page, and document templates as code with Unlayer, generate designs from prompts, import HTML, and export to HTML, PDF, or PNG.",
      "provider": "Unlayer",
      "categories": [
        "content-media",
        "code-devtools"
      ],
      "accessMethods": [
        "mcp"
      ],
      "verified": true,
      "official": false,
      "featured": false,
      "authType": "oauth2",
      "why": "Render And Export: Exports saved templates to HTML, PDF, or PNG and renders email-tuned HTML or preview images."
    },
    {
      "name": "Casetext / CoCounsel",
      "slug": "casetext-cocounsel",
      "url": "https://agentswitchboard.dev/agents/casetext-cocounsel",
      "iconUrl": "https://agentswitchboard.dev/icons/casetext-cocounsel.png?v=21e6a0d7",
      "description": "AI legal research assistant by Thomson Reuters that reviews documents, drafts memos, analyzes contracts, and finds relevant case law.",
      "provider": "Thomson Reuters",
      "categories": [
        "legal",
        "research"
      ],
      "accessMethods": [
        "api"
      ],
      "verified": true,
      "official": true,
      "featured": false,
      "authType": "apiKey",
      "why": null
    },
    {
      "name": "Unstructured",
      "slug": "unstructured",
      "url": "https://agentswitchboard.dev/agents/unstructured",
      "iconUrl": "https://agentswitchboard.dev/icons/unstructured.png?v=07b63658",
      "description": "Transforms PDFs, spreadsheets, emails, and 65+ file types into structured, chunked data ready for RAG pipelines and LLM ingestion at scale.",
      "provider": "Unstructured.io",
      "categories": [
        "data-analytics"
      ],
      "accessMethods": [
        "api",
        "mcp"
      ],
      "verified": true,
      "official": false,
      "featured": false,
      "authType": "apiKey",
      "why": null
    },
    {
      "name": "Khoj",
      "slug": "khoj",
      "url": "https://agentswitchboard.dev/agents/khoj",
      "iconUrl": "https://agentswitchboard.dev/icons/khoj.png?v=7c741eea",
      "description": "Self-hostable AI second brain that answers over your documents and the web, runs custom agents, and schedules automations. Offers an API plus Obsidian, Emacs, and desktop clients. 35k+ stars.",
      "provider": "Khoj AI",
      "categories": [
        "research"
      ],
      "accessMethods": [
        "api"
      ],
      "verified": true,
      "official": false,
      "featured": false,
      "authType": "apiKey",
      "why": "Document Q&A: Indexes personal documents and answers questions over them with citations across PDF, Markdown, and notes."
    },
    {
      "name": "LightRAG",
      "slug": "lightrag",
      "url": "https://agentswitchboard.dev/agents/lightrag",
      "iconUrl": "https://agentswitchboard.dev/icons/lightrag.png?v=519726d5",
      "description": "Builds a graph-based retrieval index over documents so agents can answer questions spanning many sources, served through a standalone API server.",
      "provider": "HKU Data Science Lab",
      "categories": [
        "memory-state",
        "research",
        "vector-databases"
      ],
      "accessMethods": [
        "api",
        "cli"
      ],
      "verified": true,
      "official": false,
      "featured": true,
      "authType": "none",
      "why": "Graph Index: Builds an entity and relationship graph from raw documents to support multi-hop retrieval."
    },
    {
      "name": "AnythingLLM",
      "slug": "anything-llm",
      "url": "https://agentswitchboard.dev/agents/anything-llm",
      "iconUrl": "https://agentswitchboard.dev/icons/anything-llm.png?v=58e38881",
      "description": "Runs an all-in-one, self-hosted RAG and agents app: chat with documents, build agents, and connect tools via MCP. Desktop and Docker with an API. By Mintplex Labs.",
      "provider": "Mintplex Labs",
      "categories": [
        "research",
        "code-devtools"
      ],
      "accessMethods": [
        "api",
        "cli",
        "mcp"
      ],
      "verified": true,
      "official": false,
      "featured": false,
      "authType": "apiKey",
      "why": "Document RAG: Ingests documents into a private workspace and answers questions over them with cited retrieval."
    },
    {
      "name": "json-render",
      "slug": "json-render",
      "url": "https://agentswitchboard.dev/agents/json-render",
      "iconUrl": "https://agentswitchboard.dev/icons/json-render.png?v=7f93989f",
      "description": "Renders AI-generated JSON specs into UIs built only from a developer-defined component catalog, with renderers for React, Vue, Svelte, React Native, PDF, and email.",
      "provider": "Vercel Labs",
      "categories": [
        "code-devtools",
        "content-media"
      ],
      "accessMethods": [
        "api"
      ],
      "verified": true,
      "official": true,
      "featured": false,
      "authType": "none",
      "why": "Cross-Platform Renderers: Renders one spec in React, Vue, Svelte, SolidJS, React Native, Ink terminals, Remotion video, PDF, or HTML email."
    },
    {
      "name": "Next AI Draw.io",
      "slug": "next-ai-draw-io",
      "url": "https://agentswitchboard.dev/agents/next-ai-draw-io",
      "iconUrl": "https://agentswitchboard.dev/icons/next-ai-draw-io.png?v=1dced7bb",
      "description": "Draws and edits draw.io diagrams from chat prompts, images, or documents for engineers and analysts, via web app, desktop app, or MCP server.",
      "provider": "Dayuan Jiang",
      "categories": [
        "content-media",
        "productivity"
      ],
      "accessMethods": [
        "mcp"
      ],
      "verified": true,
      "official": false,
      "featured": false,
      "authType": "none",
      "why": "Image and File to Diagram: Redraws an uploaded screenshot or sketch, or diagrams the content of PDF, Markdown, or code files."
    },
    {
      "name": "Sanity MCP",
      "slug": "sanity-mcp",
      "url": "https://agentswitchboard.dev/agents/sanity-mcp",
      "iconUrl": "https://agentswitchboard.dev/icons/sanity-mcp.png?v=15bdc754",
      "description": "Lets agents run GROQ queries, patch and publish documents, manage releases, and deploy schemas in Sanity projects over a hosted OAuth MCP server.",
      "provider": "Sanity",
      "categories": [
        "content-media",
        "code-devtools"
      ],
      "accessMethods": [
        "mcp"
      ],
      "verified": true,
      "official": false,
      "featured": false,
      "authType": "oauth2",
      "why": "GROQ Queries: Runs GROQ queries against a dataset with query_documents and fetches single documents with get_document."
    },
    {
      "name": "Mac Control MCP",
      "slug": "mac-control-mcp",
      "url": "https://agentswitchboard.dev/agents/mac-control-mcp",
      "iconUrl": "https://agentswitchboard.dev/icons/mac-control-mcp.png?v=91b62e46",
      "description": "Native Swift MCP server with 63 tools for full macOS automation — accessibility tree, Safari/Chrome control, screen capture, OCR, keyboard/mouse, Spotlight, and clipboard.",
      "provider": "AdelElo",
      "categories": [
        "browser-computer"
      ],
      "accessMethods": [
        "mcp"
      ],
      "verified": true,
      "official": false,
      "featured": false,
      "authType": "none",
      "why": "Screen Capture & OCR: Capture screenshots and extract text from any screen region using native macOS OCR engine"
    },
    {
      "name": "Graphify",
      "slug": "graphify",
      "url": "https://agentswitchboard.dev/agents/graphify",
      "iconUrl": "https://agentswitchboard.dev/icons/graphify.png?v=97095e6f",
      "description": "Maps a codebase, docs, PDFs, and media into a queryable knowledge graph so AI coding assistants trace and cite paths instead of grepping through files.",
      "provider": "Graphify Labs",
      "categories": [
        "code-devtools",
        "research",
        "memory-state"
      ],
      "accessMethods": [
        "cli",
        "mcp"
      ],
      "verified": true,
      "official": false,
      "featured": false,
      "authType": "none",
      "why": null
    },
    {
      "name": "LLM Wiki",
      "slug": "llm-wiki",
      "url": "https://agentswitchboard.dev/agents/llm-wiki",
      "iconUrl": "https://agentswitchboard.dev/icons/llm-wiki.png?v=84c41a45",
      "description": "Builds and maintains an interlinked wiki from your documents in a desktop app, and exposes hybrid search, graph and file reads to agents through a local HTTP API and MCP server.",
      "provider": "nashsu",
      "categories": [
        "memory-state",
        "research"
      ],
      "accessMethods": [
        "api",
        "mcp",
        "browser-extension"
      ],
      "verified": true,
      "official": false,
      "featured": false,
      "authType": "apiKey",
      "why": null
    }
  ],
  "relatedTasks": [
    "search-my-documents",
    "scrape-websites",
    "analyze-spreadsheets",
    "search-the-web"
  ],
  "stacks": [
    "research-analyst",
    "no-key-mcp-starter-pack"
  ]
}