A local document parsing tool that converts PDFs into machine-readable Markdown and JSON while preserving complex structures like formulas and tables.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install pdf-parser-mineru
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-parser-mineru 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 MinerU PDF Processor?
The MinerU PDF Processor is a high-performance local tool designed to bridge the gap between static PDF documents and structured, machine-readable data. As a key component of the Openclaw Skills library, it utilizes the MinerU engine to perform deep layout analysis, allowing users to extract text, mathematical formulas, and tables with exceptional accuracy.
This skill is particularly valuable for workflows requiring local data processing where privacy and precision are paramount. By transforming unstructured PDFs into clean Markdown or detailed JSON, it enables seamless integration with LLMs, search indexes, and version-controlled documentation systems without relying on third-party cloud services.
MinerU PDF Processor Use Cases
- Parsing academic papers to extract LaTeX-ready formulas and high-fidelity tables for research analysis.
- Converting technical documentation into Markdown for hosting on documentation sites or GitHub repositories.
- Processing scanned PDFs or documents with garbled text using advanced OCR across 109 supported languages.
- Automating batch document conversion for local RAG (Retrieval-Augmented Generation) pipelines within Openclaw Skills environments.
How MinerU PDF Processor Works
- The user provides an absolute file path and an output directory to the skill.
- The skill selects the optimal processing backend (Hybrid, Pipeline, or VLM) based on the user's hardware and accuracy needs.
- MinerU performs layout detection to identify headers, text blocks, images, and tables.
- Specialized sub-engines process complex elements like mathematical equations and multi-column layouts.
- The results are serialized into either Markdown for human readability or JSON for programmatic data analysis.
- All extracted assets, such as images and table data, are saved to the specified output directory.
MinerU PDF Processor Setup
To use this within the Openclaw Skills ecosystem, you must install the MinerU library and its dependencies locally:
# Update pip and install the uv package manager
pip install --upgrade pip
pip install uv
# Install MinerU with all feature sets
uv pip install -U "mineru[all]"
# Verify the installation
mineru --version
Ensure your system meets the requirements: Python 3.10-3.13 and at least 16GB of RAM (32GB recommended for large documents).
MinerU PDF Processor Data Schema & Taxonomy
The skill produces a structured output directory organized as follows:
| Data Component | Description |
|---|---|
| Markdown Output | A .md file containing the document text, formulas, and references to images. |
| JSON Metadata | A .json file detailing bounding boxes (bbox), page sizes, and block types. |
| Assets Folder | A directory containing all extracted images and figures from the PDF. |
| Metadata Object | Includes total page counts, author info, and document title. |
MinerU PDF Processor Advanced Features
- Support for multiple execution backends, including
vlm-auto-enginefor GPU-accelerated high-accuracy parsing. - Granular page control allowing users to process specific ranges (start_page/end_page) to save time and memory.
- Built-in multilingual OCR support for over 100 languages, automatically detectable or manually specifiable.
- Capability to toggle heavy processing features like formula recognition or table extraction for faster performance on simple documents.
SKILL.md
Loading
Related Openclaw Skills
GLM-OCR Multimodal Document Deep Analysis
A high-precision document parsing pipeline that extracts layout elements and performs deep multimodal analysis on tables and charts.
Feishu Leave Request
A browser automation skill that facilitates the end-to-end submission of leave requests within the Feishu (Lark) corporate ecosystem.
Trading Analyzer
A unified market analysis engine for cryptocurrency and stocks powered by multi-source financial data and technical indicators.
byt-workflow
A sophisticated automation pipeline that downloads YouTube audio and captures real-time AI translations using the Doubao interface.
Turkiye AFAD Deprem Skill
A specialized tool for retrieving and filtering Turkish earthquake data from official AFAD sources.
ClawShot
ClawShot is the visual social network for AI agents to showcase their work, build a following, and engage through likes, comments, and mentions.