skills.rest
galeep
galeepCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

liteparse

Extract text and metadata from PDFs, DOCX, and images with spatial bounding boxes.

Updated May 17, 2026
Search Tags:#ocr#metadata extraction#document parsing#local processing#pdf text extraction
One-click install
npx skills add https://github.com/galeep/plugin-place --skill liteparse-galeep
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: liteparse
Source: https://github.com/galeep/plugin-place/tree/main/plugins/sci-scientific-communication/skills/liteparse
Command: npx skills add https://github.com/galeep/plugin-place --skill liteparse-galeep

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires liteparse, libreoffice, imagemagick, and includes scripts (resource) and references (resource) components.

What problem does it solve?

LiteParse addresses the challenge of efficiently extracting text and metadata from a variety of document formats, including PDFs, DOCX, and images, without relying on cloud services.

Core Features & Use Cases

  • Layout-Aware Text Extraction: Provides spatial information with text, allowing for tasks like layout-aware search and reference grounding.
  • Fast Local Parsing: Enables quick text extraction and conversion of PDFs and office documents into structured JSON for easy integration.
  • OCR: Supports OCR on scanned PDFs and images for additional text extraction from non-textual documents.
  • Use Case: Imagine you need to quickly extract and analyze the text content from multiple scientific papers. LiteParse can parse and extract the text, allowing for easy processing and analysis.

Quick Start

Run the 'lit parse' command with the path to your document to begin parsing.

Frequently Asked Questions about liteparse

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I extract text and metadata from PDFs and DOCX locally without cloud APIs?▼

Local PDF text extraction and metadata parsing can be achieved using local Python libraries to convert documents into structured JSON. This approach processes files entirely offline without relying on cloud services, ensuring data privacy and fast local parsing.

Can I get spatial bounding boxes with extracted text for layout-aware search?▼

Yes, layout-aware text extraction provides spatial bounding boxes alongside the extracted text. This spatial information enables precise layout-aware search and reference grounding directly from your parsed PDFs and office documents.

Does local PDF text extraction support scanned documents and image files?▼

Local PDF text extraction includes OCR capabilities to handle scanned PDFs and image files. This allows you to perform text extraction from non-textual documents and images, extending parsing coverage to digitized literature.

Do I need LibreOffice and ImageMagick installed to parse office documents?▼

Yes, full functionality for local document parsing requires installing Python, LibreOffice, and ImageMagick. These dependencies enable the conversion and processing of DOCX files and image formats alongside standard PDF text extraction.

What is the best way to batch ingest text from multiple scientific papers?▼

Batch ingestion of literature folders is supported by running the parse command with your directory path. This allows you to quickly extract and analyze text content from multiple scientific papers, outputting structured JSON for easy integration.

Related Skills

Explore More Data & Analytics Skills→
Ritabrata-ChakrabortyRitabrata-Chakraborty

liteparse

Extract spatial text and bounding boxes from local documents with OCR.

Community
Advanced
antonyfmunozantonyfmunoz

x_twitter_api

Search X/Twitter posts by keyword and handle with normalized JSON output.

Community
Intermediate
SignmanalSignmanal
1

chroma

Store and query vector embeddings with metadata filtering for RAG applications.

Community
Intermediate
anxiety135790anxiety135790

ocr-and-documents

Extract text from PDFs, scanned documents, and images into plain text or markdown.

Community
Advanced