Skill Detail
Convert dense PDFs into LLM-ready text and page-aligned markdown with olmOCR
Use olmOCR when an agent needs to turn scanned or layout-heavy documents into clean markdown or text before chunking, search, extraction, or citation workflows.
Data Extraction & TransformationMulti-Framework
Data Extraction & Transformation
Multi-Framework
Security Reviewed
⭐ 17.1k GitHub stars
INSTALL WITH ANY AGENT
npx skills add agentskillexchange/skills --skill convert-dense-pdfs-into-llm-ready-text-and-page-aligned-markdown-with-olmocr
Works best when you want a reusable capability, not another fragile one-off prompt.
At a glance
Tools required
Python 3.11, pip or conda, poppler-utils, optional NVIDIA GPU for local inference
Install & setup
Create a clean Python environment, install required PDF rendering dependencies, then install the package with pip install olmocr for remote inference or pip install olmocr[gpu] for local GPU inference.
Author
Allen Institute for AI
Publisher
Organization
Last updated
Apr 16, 2026
Quick brief
Use olmOCR when the job is to convert PDFs or image-based documents into readable markdown or plain text with natural reading order, table handling, and header or footer cleanup. Invoke it instead of treating the original PDF as the working surface when downstream agent steps depend on clean text for retrieval, extraction, QA, or citation. The scope boundary is specific and skill-shaped: this is a document linearization and OCR preprocessing workflow, not a general document platform listing and not just a raw OCR model card.