A specialized toolkit for local PDF processing, including OCR for scanned documents and automated arXiv reference extraction.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install pdf-utils
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-utils 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 Utils?
PDF Utils is a high-performance technical extension designed for Openclaw Skills that bridges the gap between simple PDF reading and complex document engineering. While standard tools handle semantic analysis, this skill provides a robust framework for local, scriptable operations using PyMuPDF and Tesseract. It is specifically optimized for developers and researchers who need to handle non-searchable, image-based PDFs or automate the extraction of scientific metadata.
By integrating these Openclaw Skills into your workflow, you gain programmatic control over document manipulation. Whether you are dealing with legacy scanned papers or need to batch-process academic references, PDF Utils offers a stable 1.x environment to ensure high-fidelity data extraction without relying on external cloud-based parsing services.
PDF Utils Use Cases
- Processing scanned or image-based PDF documents where standard text extraction fails.
- Automated mining of arXiv IDs from research papers for literature reviews.
- Batch downloading referenced academic papers based on extracted IDs.
- Performing repeatable local operations like merging, splitting, or rendering PDF pages to images.
- Improving AI agent reliability when handling low-resolution or handwritten document scans.
How PDF Utils Works
- The user provides a target PDF file path to the specific utility script within the Openclaw Skills directory.
- For text-based PDFs, the extraction script parses the internal metadata and text layer to identify arXiv patterns.
- For scanned documents, the OCR engine (Tesseract) processes the page images to synthesize a searchable text layer.
- The system optionally triggers downstream actions, such as fetching full-text PDFs from the arXiv API.
- Results are returned as structured text output or saved as new files according to the user's CLI flags.
PDF Utils Setup
To utilize these features within your Openclaw Skills environment, ensure the following system dependencies are installed:
# Install system OCR engines
brew install tesseract
brew install tesseract-lang
# Install Python dependencies
pip3 install pytesseract Pillow pymupdf --break-system-packages
PDF Utils Data Schema & Taxonomy
The skill manages data through a combination of CLI arguments and local file outputs:
| Script | Input | Primary Output |
|---|---|---|
extract_refs.py |
.pdf file | arXiv IDs, downloaded PDFs |
ocr_pdf.py |
.pdf (scanned) | OCR-processed text/extracted refs |
pdf_ops.py |
.pdf file(s) | Merged/Split files, page images |
Metadata is processed in-memory during script execution and can be directed to local directories using the --out flag.
PDF Utils Advanced Features
- Multi-page OCR support with the
--allflag for comprehensive document scanning. - Integrated reference mining that combines OCR and extraction in a single pipeline step.
- Granular page range processing to optimize performance on large-scale documents.
- Programmatic API access for integrating PDF operations into custom Openclaw Skills workflows.
- Local rendering engine to convert specific PDF pages into high-quality images for visual analysis.
SKILL.md
Loading
Related Openclaw Skills
Dispatching Parallel Agents
A sophisticated workflow pattern for delegating independent tasks to specialized AI agents with isolated contexts to enable concurrent problem-solving.
Financial Analyzer
An AI-powered financial assistant for deep statement analysis, investment valuation, and automated risk assessment.
Bing Image Search MCP
A specialized skill enabling AI agents to programmatically search and retrieve image data from Bing via the OneKey Gateway.
hermit-search
A local semantic search service providing hybrid retrieval and cross-encoder reranking for multi-collection knowledge bases.
OpenCLI
OpenCLI is a versatile bridge that allows AI agents to interact with web platforms, desktop applications, and system-level CLI tools through a unified interface.
Gumroad Page Generator
The Gumroad Page Generator automates the creation of high-converting product page content based on specific audience and feature inputs.
Featured*
