A powerful OCR tool that utilizes the MiniMax Vision API to convert PDF documents and images into structured Markdown text.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install minimax-pdf-ocr
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 minimax-pdf-ocr 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 MiniMax PDF OCR?
The MiniMax PDF OCR skill is a specialized tool within the Openclaw Skills ecosystem designed for high-fidelity text extraction from complex visual documents. By leveraging the MiniMax M2.5 Vision model, it goes beyond simple character recognition to provide nuanced visual understanding of document layouts.
This skill is particularly effective for developers needing to digitize physical documents, research papers, or images into clean, editable Markdown format. It handles the heavy lifting of converting PDF pages into images via poppler before processing them through the MiniMax API to ensure the highest possible recognition accuracy for both Chinese and English text.
MiniMax PDF OCR Use Cases
- Digitizing printed research papers and reports into searchable Markdown files.
- Automating the extraction of text from scanned legal or administrative PDF documents.
- Integrating high-accuracy OCR capabilities into AI agent workflows using Openclaw Skills.
- Converting image-heavy documentation into structured text for LLM fine-tuning or RAG systems.
How MiniMax PDF OCR Works
- The skill takes a PDF file path and an optional output directory as input.
- It utilizes the poppler library to render each PDF page into a high-resolution image.
- Each generated image is sent to the MiniMax Vision API for deep visual analysis and text extraction.
- The skill processes the API response to maintain original paragraph structures and formatting.
- All extracted text is compiled and saved as a comprehensive .md file in the specified output directory.
MiniMax PDF OCR Setup
To get started with this component of Openclaw Skills, ensure you have poppler installed on your system along with the necessary Node.js dependencies:
# Install system dependencies (macOS)
brew install poppler
# Navigate to skill directory and install Node.js packages
cd minimax-pdf-ocr
npm install openai pdf2image
# Set your API credentials
export MINIMAX_API_KEY="your-api-key"
MiniMax PDF OCR Data Schema & Taxonomy
The skill organizes processed data into a structured output format to ensure compatibility with other Openclaw Skills. The output is typically a Markdown file with the following taxonomy:
| Feature | Description |
|---|---|
| File Format | Standard Markdown (.md) |
| Header Structure | Level 1 header for document title; Level 2 headers for individual page markers |
| Content | Preserved paragraph breaks and structural flow from the original document |
| Metadata | Integrated page numbering for easy reference |
MiniMax PDF OCR Advanced Features
- Multi-language support optimized for high-precision Chinese and English recognition.
- Programmatic integration allowing the
recognizePdffunction to be called within larger JavaScript automation scripts. - Native support for the MiniMax M2.5 model, providing state-of-the-art visual understanding and layout analysis.
- Scalable architecture suitable for batch processing by implementing custom delays to respect API rate limits.
SKILL.md
Loading
Related Openclaw Skills
Auto-Data-Analyzer
An automated data analysis tool that transforms raw CSV and Excel files into comprehensive HTML reports with cleaning, statistics, and visualizations.
Pixel Familiar Identity
A comprehensive AI agent identity and SOUL.md framework designed to power revenue-focused autonomous business agents.
AgentMemory Skill
A persistent memory framework that allows AI agents to remember facts, track entities, and learn from past experiences across sessions.
Email Sequence Skill
An AI-powered copywriting engine designed to architect and write multi-stage email marketing campaigns with strategic timing and mobile-first formatting.
agent-browser-zh
A high-performance Rust-based headless browser automation CLI designed for AI agents to interact with the web using structured commands.
Claw Browser Automation
A comprehensive CLI-based browser automation toolkit designed for AI agents to navigate, interact, and extract data from web applications.
Featured*
