pymupdf-pdf

What This Skill Does

The pymupdf-pdf skill provides a high-performance, local PDF extraction interface for the OpenClaw agent. Utilizing the PyMuPDF (fitz) library, it enables the rapid conversion of PDF documents into clean Markdown or structured JSON formats. Unlike heavy-duty OCR tools, this skill is optimized for speed and efficiency, making it the ideal choice for processing large volumes of standard documents or for quick-look document ingestion when latency is a primary concern. It supports granular features like optional image extraction and basic table parsing, outputting files into organized, per-document directories to maintain workspace hygiene.

Installation

To integrate this skill into your OpenClaw environment, execute the following command in your terminal: clawhub install openclaw/skills/skills/kesslerio/pymupdf-pdf-parser-clawdbot-skill

Ensure that your system environment is configured with the necessary dependencies. If you encounter errors related to missing libraries or Nix-specific linking issues, please consult the documentation provided in references/pymupdf-notes.md within the skill source repository.

Use Cases

This skill is best deployed in scenarios where document processing speed is prioritized over complex layout reconstruction. Use cases include:

• Rapidly scanning internal documentation or manuals to populate an agent's knowledge base.
• Bulk-processing PDF archives where high-fidelity OCR is unnecessary.
• Serving as a secondary, fallback parser when heavier OCR systems (like MinerU) are currently unavailable or failing on simple documents.
• Extracting textual content from clean, machine-generated PDFs to facilitate downstream RAG (Retrieval-Augmented Generation) workflows.

Example Prompts

1. "Use the pymupdf-pdf skill to parse the quarterly_report.pdf file and output the results in Markdown format to my ./parsed-docs directory."
2. "Extract all text and images from contract_final.pdf using the pymupdf-pdf parser."
3. "Run a quick parse on the technical-manual.pdf, generate both Markdown and JSON outputs, and include table extraction to help me identify the data structure."

Tips & Limitations

• Speed vs. Robustness: PyMuPDF is extremely fast but lacks the complex multi-column layout analysis of more advanced OCR tools. For documents with complex headers, multi-column layouts, or heavy formatting, expect occasional structural errors.
• Image/Table Quality: Image extraction is straightforward, but table parsing is limited to line-based estimation; do not rely on it for complex financial spreadsheets.
• Organization: By default, the skill creates organized subdirectories. Avoid changing the --outroot to a root directory unless you have a specific reason to bypass this automatic cleanup.
• Troubleshooting: If the parser struggles with a specific file, check the PDF for security permissions or non-standard encoding, then consider falling back to a more intensive OCR tool if required.