Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Builds HTML and PDF slide decks from a markdown file, URL, or topic description. Generates AI illustrations per slide, validates each slide against design reference images using Playwright screenshots, and exports a content.md file for future edits.
Combines content drafting, AI illustration, design-reference matching, per-slide screenshot validation, and PDF export in a single automated workflow rather than requiring separate tools for each step.
When to use it
- Turning a blog post URL into a conference talk deck
- Converting a project README into a stakeholder presentation
- Generating a branded slide deck for a new product pitch
- Iterating on slides by editing content.md and regenerating
- Creating a PDF handout from an unstructured topic description
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Generate a presentation about Kubernetes networking from https://kubernetes.io/docs/concepts/cluster-administration/networking/
- 1Fetch URL content with WebFetch and extract key concepts
- 2Write presentation/content.md draft and wait for user approval
- 3Read reference images in references/ to extract color palette, typography, and layout patterns
- 4Generate presentation/slides.html with inline styles matching the reference design
- 5Call mcp__openai-gpt-image-mcp__create-image for each slide to produce matching illustrations and embed them
- 6Open slides.html in Playwright at 1920x1080, screenshot each slide, compare against references, and fix mismatches
- 7Run slides_to_pdf.py to combine slide_N.png files into presentation/presentation.pdf
presentation/slides.html (interactive), presentation/presentation.pdf, presentation/slide_N.png per slide, presentation/content.md for future edits
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.