12 tools
Content-analysis toolkit that detects file types and extracts text and metadata from over a thousand file formats (including PDF, DOCX, PPTX, XLSX) through a single parsing interface; supports OCR (via Tesseract), language detection, named-entity recognition, and parsing of embedded documents/attachments. Available as a Java library, CLI (tika-app), or REST server (tika-server).
Docker-packaged drop-in alternative to wkhtmltopdf: a CLI (`athenapdf`, built on Electron) for direct HTML-to-PDF conversion with an 'aggressive mode' that strips page clutter like ads and navigation before rendering. Also ships a `weaver` HTTP microservice (written in Go) for running conversions as a scalable, horizontally-deployable service on Kubernetes or AWS ECS. Built on Electron/Chromium rather than the discontinued Qt WebKit engine wkhtmltopdf itself relies on.
Go library, CLI, and HTTP service that converts PDF, DOC, DOCX, XML, HTML, RTF, ODT, Pages, and image files to plain text plus metadata (as JSON), aiming to be a single conversion entry point across many document formats rather than a PDF-specific tool. Optional OCR support (via Tesseract, enabled with the `ocr` build tag) handles image-based input. Its multi-format scope suits ingestion pipelines that need to normalize mixed document types into plain text.
Node.js web service (not a library — you deploy or call an already-running instance) that converts HTML into PDF, with the HTML passed as a data URI in the request rather than a file upload. Supports manual page breaks via the CSS `page-break-before: always` rule, giving callers explicit control over pagination rather than relying purely on automatic content-flow breaking. A minimal, single-endpoint alternative to running a full headless-browser rendering pipeline yourself.
Self-hosted PDF layout-analysis service (choice of a VGT or LightGBM model) exposed over an HTTP API: segments and classifies page regions — text, titles, pictures, tables, and similar — as structured output rather than a flat text dump. Built by HURIDOCS, a human-rights documentation NGO, originally for processing large volumes of case files where a region's semantic type (not just its text) matters for downstream analysis. Runs as a Docker service you deploy and call, not a library you import.
RubyPure-Ruby (no external system libraries) invoice-PDF generator, usable as a library, CLI, or server: A4/US Letter formats, purchaser/provider info boxes with addresses and tax IDs, SWIFT/IBAN payment details, itemized tables (description/quantity/price/tax/totals), automatic page numbering, customizable labels, custom fonts/logos/backgrounds/stamps/signatures/QR codes, and JSON-driven invoice data — fast enough for real-time request handling.
PythonConverts PDF, DOCX, PPTX, XLSX, images, and web pages into Markdown or JSON: strips headers/footers/page numbers while preserving structure, converts formulas to LaTeX and tables to HTML, OCRs scanned/garbled PDFs in 109 languages, handles single/multi-column and complex layouts (incl. cross-page table merging), and extracts images with captions. Offers CLI, FastAPI, and Gradio WebUI, plus Python/Go/TypeScript SDKs; builds on PDF-Extract-Kit's models.
PythonApache-2.0 self-hosted document-intelligence platform: manages document collections, automatically parses PDF layout and generates vector embeddings, provides a pluggable microservice analyzer architecture for automated annotation, and a human annotation interface (multi-page annotations) built on the PAWLS annotation format. Exposes a GraphQL+REST API, a Model Context Protocol server for agents, and a React UI.
PythonWeb-based tool (built by AllenAI) for collecting human annotations on PDFs, purpose-built for academic papers but usable on any PDF collection: preprocesses PDFs into per-page token bounding boxes (optionally via GROBID), serves them in a browser viewer, assigns documents to specific annotators, enforces login/access control, and stores annotations via an API backend.
PDF toolkit built around a Rust engine, exposed as native bindings for ~20 languages (Python, JavaScript/TypeScript, Go, C#/.NET, Java, and more), a CLI, and an MCP server for AI-assistant integration: text/character-level and image extraction, form-field reading and filling, Markdown/HTML conversion with heading detection, PDF creation and editing, watermarking, merging/splitting, regex-based search, and metadata access.
Self-hostable enterprise search and agentic-workflow platform that unifies business data (PDF, Word, Excel, PowerPoint, and more) into a searchable knowledge layer: parses PDFs via pdfplumber, with optional Docling-based multimodal parsing and LLM-based OCR for scanned documents.
JavaScriptMicroservice (Puppeteer/headless Chrome) that converts URLs or posted HTML into PDFs or PNG/JPEG screenshots, with configurable viewport, PDF page size/orientation/margins/scale, header/footer templates, page-range selection, cookies/headers, SPA network-idle waiting, and optional API-key auth; deployable via Heroku or Docker.