Document organizer
[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
npx -y skills add cxcscmu/SkillLearnBench --skill document-organizerAssembled 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
Organizes mixed document files (PDF, PPTX, DOCX) into subject-based folders by analyzing content. Use this skill when sorting, categorizing, or organizing documents into topic folders.
SKILL.md
1.4 KB, as published. Nobody here has run it
Document Organizer
Sorts documents of various formats into predefined subject folders based on content analysis.
Workflow
- Inventory all files in the source directory
- Extract text from each file format:
- PDF: use PyPDF2 to extract text from first pages
- DOCX: use python-docx to read paragraphs
- PPTX: use python-pptx to read slide text
- Classify each file using keyword-based scoring (see pdf-classifier skill)
- Move files to target folders using shutil.move()
- Verify all files are moved, none left behind
Target Folder Structure
papers/
├── LLM/
├── trapped_ion_and_qc/
├── black_hole/
├── DNA/
└── music_history/
Key Principles
- Never rename files — preserve original filenames
- Never modify file content
- Each file goes to exactly one folder
- If a file doesn't clearly match any of the first 4 categories, assign it to
music_history(the catch-all) - Process all file types (PDF, DOCX, PPTX) with format-appropriate text extraction
Python Dependencies
PyPDF2for PDFspython-docxfor DOCX filespython-pptxfor PPTX files- All available via pip if not already installed