Pdf text extraction

Skill cxcscmu/SkillLearnBench/skills/b1-one-shot-claude-haiku-4-5/organize-messy-files/pdf-text-extraction

[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.

Install

npx -y skills add cxcscmu/SkillLearnBench --skill pdf-text-extraction

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

What its author says it does

Copied from the file, not written here

Extract text content from PDF files for analysis and classification

SKILL.md

1.6 KB, as published. Nobody here has run it

PDF Text Extraction

Overview

Extract text content from PDF files to analyze their content and classify them by subject.

Installation

pip install pdfplumber PyPDF2

Usage Examples

Using pdfplumber (Recommended)

import pdfplumber

def extract_pdf_text(pdf_path, max_chars=5000):
    """Extract text from PDF with character limit for efficiency"""
    try:
        with pdfplumber.open(pdf_path) as pdf:
            text = ""
            # Read first few pages to get representative content
            for page_num in range(min(3, len(pdf.pages))):
                text += pdf.pages[page_num].extract_text() or ""
                if len(text) > max_chars:
                    break
            return text[:max_chars]
    except Exception as e:
        return f"Error reading PDF: {str(e)}"

Using PyPDF2 (Fallback)

from PyPDF2 import PdfReader

def extract_pdf_text_pypdf(pdf_path):
    """Alternative PDF text extraction"""
    try:
        reader = PdfReader(pdf_path)
        text = ""
        for page in reader.pages[:3]:  # First 3 pages
            text += page.extract_text()
        return text
    except Exception as e:
        return f"Error: {str(e)}"

Best Practices

Extract from first 2-3 pages only (faster, usually contains abstracts/titles)
Handle errors gracefully for corrupted PDFs
Cache extracted text to avoid re-processing
Use reasonable character limits (3000-5000 chars) for classification