Pptx

What it does

This skill enables an AI agent to create, edit, and analyze PowerPoint (.pptx) files. A .pptx file is a ZIP archive containing XML files and other resources. The skill provides tools for extracting text content, accessing raw XML data (for elements like comments, speaker notes, animations, and formatting), and analyzing design elements such as typography and colors. It also supports creating new presentations using an HTML-to-PowerPoint conversion workflow.

When to use it

• You need to extract the text from a PowerPoint presentation.
• You require access to specific elements within a .pptx file, like speaker notes or slide layouts.
• You want to analyze the design (colors and fonts) of an existing PowerPoint presentation as inspiration for a new one.
• You need to create a new PowerPoint presentation from scratch, leveraging HTML content.
• You need to understand the structure of a .pptx file for debugging or advanced manipulation.

Key capabilities

• Text Extraction: Converts .pptx files to Markdown format.
• Raw XML Access: Allows access to comments, speaker notes, slide layouts, animations, design elements, and complex formatting within the .pptx file.
• Typography & Color Analysis: Extracts color schemes ( <a:clrScheme>) and font information (<a:fontScheme>) from theme files, and analyzes actual font usage in slides.
• File Structure Understanding: Provides knowledge of key file structures within a .pptx archive (e.g., presentation.xml, slide{N}.xml, theme/theme1.xml).
• HTML to PowerPoint Conversion: Creates presentations from HTML content using the html2pptx workflow.

Example prompts

• "Extract all the text from this PowerPoint file: [path-to-file.pptx]"
• "What fonts and colors are used in this presentation? [path-to-file.pptx]"
• "Create a new PowerPoint presentation based on this HTML content."
• "Show me the speaker notes for slide 3 of this presentation: [path-to-file.pptx]"

Tips & gotchas

• Design Approach First: Always state your design approach (tone, branding, color palette) before generating any code or creating a presentation.
• Web-Safe Fonts Only: The skill is designed to use only web-safe fonts: Arial, Helvetica, Times New Roman, Georgia, Courier New, Verdana, Tahoma, Trebuchet MS, Impact.
• Script Location: If using the unpack.py script, be aware that its location might vary and may require locating it with find . -name "unpack.py".