Back to feed
Dev.to
Dev.to
6/24/2026
Using LLMs for slide: MDPR + mdpr-skill

Using LLMs for slide: MDPR + mdpr-skill

Short summary

MDPR is a Markdown-to-PowerPoint runtime that solves LLM slide generation's core problem: models often own coordinates, colors, and spacing, causing visual drift. The solution separates concerns—LLMs suggest structure via mdpr-skill, while MDPR deterministically handles layout, theming, and rendering. Output is editable PPTX with HTML/PDF previews; creator seeks feedback on edge cases like dense tables and multilingual content.

  • •MDPR separates LLM suggestions from deterministic presentation rendering to prevent visual drift across decks
  • •mdpr-skill provides AI review hints without owning final coordinates, colors, or geometry
  • •Outputs editable PPTX with HTML/PDF previews; install via npm

Generated with AI, which can make mistakes.

#ai-tools#open-source
Read full article at Dev.to

Is this a good recommendation for you?

Comments

Failed to load comments. Please try again.

Explore more

What I learned building a slide-deck skill for AI agents
Dev.to
The original title is "Lessons from Embedding an LLM Inside a Drag-and-Drop Editor"
Dev.to
MemSlides: A Hierarchical Memory Driven Agent Framework for Personalized Slide Generation with Multi-turn Local Revision
arXiv cs.CL
Slide Builder: RevealJS in YAML, with Prezi-style zoom
Dev.to