A specialized tool for extracting text and complex LaTeX mathematical formulas from academic papers in English and Chinese.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install pdf-extractor-skill
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install pdf-extractor-skill using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is PDF Extractor Skill?
The PDF Extractor Skill is a powerful utility designed for researchers and developers who need to convert complex academic documents into structured Markdown format. By leveraging state-of-the-art models like Marker and Nougat, this skill accurately identifies text, handles multi-column layouts, and translates mathematical notation into clean LaTeX code. It is an essential component of the Openclaw Skills ecosystem for anyone performing high-fidelity document digitisation.
This skill is particularly optimized for academic papers, including those from arXiv, and supports both English and Chinese content. Whether you are dealing with scanned PDFs that require robust OCR or digital papers with intricate formulaic structures, this skill provides a self-contained environment to automate the extraction process effectively.
PDF Extractor Skill Use Cases
- Converting academic papers into Markdown for easier reading and note-taking.
- Extracting complex LaTeX formulas from PDFs for use in research or technical documentation.
- Digitizing Chinese-language technical documents with high accuracy.
- Batch processing arXiv papers to create datasets for LLM training or retrieval-augmented generation.
How PDF Extractor Skill Works
- The user provides a PDF file path and specifies the primary language (English or Chinese).
- The system selects the optimal tool: Marker for Chinese/mixed content or Nougat for pure English papers.
- The document is processed through specialized layout analysis and OCR models (like Surya or Texify).
- Text is extracted as regular Markdown, while formulas are encapsulated in $inline$ or $$display$$ LaTeX blocks.
- Images and tables are identified and preserved, with Marker saving images into a dedicated output folder.
- The final output is generated as a structured Markdown file ready for use in PKM tools or LLMs.
PDF Extractor Skill Setup
This skill requires a dedicated Conda environment. Ensure you have CUDA 12.8 installed for optimal GPU performance.
# Navigate to the skill directory
cd C:\Users\cr\.config\opencode\skills\pdf-extractor
# Execution command using the dedicated environment
D:\anaconda3\envs\pdf-extractor\python.exe scripts\pdf2md_marker.py "your_paper.pdf" --ark-code-latest
PDF Extractor Skill Data Schema & Taxonomy
The skill organizes extracted content into structured Markdown with the following metadata and file taxonomy:
| Data Type | Format | Notes |
|---|---|---|
| Text Content | Markdown | Standard syntax with headers and lists preserved. |
| Math Formulas | LaTeX | Wrapped in $...$ or $$...$$ syntax. |
| Tables | GFM Tables | Extracted into Markdown table blocks. |
| Images | PNG/JPG | Extracted by Marker into the output directory. |
| Metadata | JSON/Header | Includes page range, language used, and model version. |
PDF Extractor Skill Advanced Features
- Multi-model routing: Automatically switches between Marker and Nougat based on language requirements.
- Per-page extraction: Support for
--page-rangeto handle extremely long documents without memory issues. - LLM Enhancement: Integration with ark-code-latest to improve extraction quality for complex table structures.
- Force OCR mode: Capability to process image-only or poorly encoded PDFs using specialized OCR models.
- Custom Language Support: Support for Japanese and other languages via the
--languagesflag in Marker.
SKILL.md
Loading
Related Openclaw Skills
Taiko Native Bridge
A professional-grade bridge interface for programmatically moving ETH and various token standards between Taiko L1 and L2 layers.
B2B SaaS Email Sequence Builder
A specialized framework for building high-converting B2B SaaS email sequences, drip campaigns, and multi-touch nurture flows.
ABM Campaign Builder
A professional-grade B2B SaaS framework for designing, tiering, and executing precision Account-Based Marketing campaigns.
B2B SaaS Positioning & Messaging Framework
A comprehensive framework for B2B SaaS companies to define their product positioning, value propositions, and differentiated messaging across multiple audiences.
Supabase Dashboard Builder
A lightweight toolkit for building framework-free admin dashboards and data visualizations powered by the Supabase REST API.
Machine Config Migrator
A low-risk, component-based migration tool for transferring development environments and productivity configurations across workstations.
Featured*
