Extract text, tables, metadata, and images from 75+ file formats.
npx skills add https://github.com/kreuzberg-dev/kreuzberg --skill kreuzberg
Please help me install this Agent Skill. Skill: kreuzberg Source: https://github.com/kreuzberg-dev/kreuzberg/tree/main/skills/kreuzberg Command: npx skills add https://github.com/kreuzberg-dev/kreuzberg --skill kreuzberg
Kreuzberg is a high-performance document intelligence library that enables programmatic extraction of text, tables, metadata, and images from 75+ formats across PDFs, Office documents, images (with OCR), HTML, email, archives, and academic formats.
Process a sample document to extract structured content and inspect results.
High-intent search queries and answers about installing and using this skill.
Batch document extraction pulls text, tables, metadata, and images from 75+ file formats in a single run. Kreuzberg supports processing multiple files simultaneously with configurable options, enabling efficient indexing and downstream analysis across Python, Node.js, Rust, and CLI environments.
OCR enables text extraction from scanned images and image-based PDFs via configurable backends. Kreuzberg automatically recognizes text in image files, making previously inaccessible scanned content available for search indexing and downstream document processing workflows.
Multi-format extraction from Office documents, HTML, email, and academic formats provides structured text and metadata ideal for search indexing. Kreuzberg handles diverse document types programmatically, enabling robust document intelligence pipelines without manual format-specific preprocessing.
Language detection and embeddings support document chunking after text extraction. Kreuzberg's extraction pipeline includes these capabilities to enable downstream analysis, ensuring extracted content from 75+ formats is properly segmented and processed for indexing workflows.
Plugin architecture supports custom OCR backends, post-processors, and validators for document extraction. Kreuzberg provides extension points that allow developers to customize the extraction pipeline, adding validation logic and specialized processing for document-heavy workflows.
Extraction supports 75+ formats including PDFs, Office documents, images, HTML, email, archives, and academic formats. Kreuzberg extracts metadata and images from these diverse file types, enabling comprehensive document intelligence across mixed-format collections.
xberg-ioExtract text, tables, and metadata from PDFs using OCR.
vecearBuild document embedding and retrieval pipelines for scalable semantic search.
beyondkmpAutomate PDF form analysis, filling, table extraction, and OCR with pypdf.
olostepScrape up to 10,000 URLs in parallel with configurable output formats.
