Dangling Paper Title Processor for PPTX
Overview
This skill processes PowerPoint presentations to find "dangling paper titles" - text boxes (non-placeholder shapes) containing paper title text. It reformats them, repositions them, and creates a summary reference slide.
What are Dangling Paper Titles?
In academic presentation slides, paper titles are often placed as free-floating text boxes (not in standard placeholders like Title or Content). These are detected by finding non-placeholder shapes with text frames that contain text.
Operations
- Detect: Find all TEXT_BOX shapes (non-placeholders) with text content
- Reformat: Change font to Arial, size 16pt, color #989596, disable bold
- Resize: Adjust box width so title displays in one line
- Reposition: Center horizontally at bottom of slide
- Reference Slide: Create new slide with "Reference" title and auto-numbered unique titles
Usage
pythonimport sys sys.path.insert(0, '/app/environment/skills/evo-pptx-dangling-titles/scripts') from utils import process_pptx, validate_output # Process the PPTX result = process_pptx('/root/Awesome-Agent-Papers.pptx', '/root/Awesome-Agent-Papers_processed.pptx') # Validate the output errors = validate_output('/root/Awesome-Agent-Papers_processed.pptx') if errors: print("VALIDATION FAILED") else: print("ALL CHECKS PASSED")
Key Functions
detect_dangling_titles(prs)- Returns list of dicts with slide_index, shape, textestimate_text_width_emu(text, font_size_pt, font_name)- Estimates EMU width for single-line textformat_dangling_title(shape, slide_width, slide_height)- Applies formatting and positioningcollect_unique_titles(dangling_items)- Deduplicates titles preserving ordercreate_reference_slide(prs, titles)- Creates reference slide with numbered listprocess_pptx(input_path, output_path)- End-to-end entry pointvalidate_output(output_path)- Validates the processed file


