LearnPrompt

LearnPrompt / humanize-ppt

Public

AST-based outline director for human-centered AI presentation workflows.

learnprompt.github.iohumanize-ppt ai-workflow ast html-slides humanizer ppt
16
4
89% credibility
Found May 20, 2026 at 18 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

Humanize PPT is a presentation planning tool that transforms raw content into audience-focused presentations. Instead of generating slides directly from notes, it first analyzes who will watch the presentation and creates a structured 'deck brief' that maps out how each slide moves the audience from confusion to understanding. The tool then generates actual slides in multiple visual styles, lets you preview and choose your favorite look, adds speaker notes with talking points, and exports a finished presentation ready to present or share.

How It Works

1
📄 You have your raw material

You have a draft, notes, article, or research you want to turn into a presentation that actually connects with your audience.

2
✨ The tool listens to your content first

Instead of jumping straight to slides, the tool reads your material and figures out who your audience is and what they need to understand.

3
🗺️ You get a clear presentation map

The tool creates a structured plan showing exactly how each slide moves your audience from confusion to clarity.

4
You choose your visual style
📝
Clean & Professional

A calm, readable style perfect for business or educational presentations

📰
Bold Editorial

A sophisticated magazine-style look for impactful storytelling

⚡
High Contrast

A dramatic, modern style that grabs attention immediately

5
👁️ You preview before committing

See how your presentation looks in your chosen style, with real slides and real content—not just empty templates.

6
🎤 You add speaker notes

The tool generates talking points for each slide so you know exactly what to say and what to avoid.

🎉 Your presentation is ready to share

Export your finished deck as a portable HTML file or PDF, complete with speaker notes and presenter controls.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 18 to 16 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is humanize-ppt?

Humanize PPT is a Python command-line tool that transforms raw content into structured presentation briefs before any slides are rendered. Instead of feeding documents directly into a slide generator, it uses AST theory to analyze your material and produce a "deck brief" covering audience state, cognitive tension, and narrative flow. The output includes slide plans, speaker notes, asset manifests, and routing decisions that downstream renderers consume.

The workflow follows an OPC pattern: Outline, Produce, Complete. You pass in a markdown file or existing PPTX, and it generates contracts and commands that specialized renderers then execute.

Why is it gaining traction?

The insight here is sharp: AI-generated slides inherit the noise from model reasoning traces and bloated summaries. This tool intervenes before rendering, cleaning the material into an audience-aware path. It supports multiple renderers including guizang and beautiful-html-templates, with adapters for presenter mode and PDF export. The auto-clone feature for template repositories lowers friction.

Who should use this?

Content creators building AI tools who want presentations that actually communicate, not just display. Developers working on PPT agents who need a canonical preprocessing layer. Product teams tired of decks that read like model outputs rather than intentional narratives.

Verdict

The 0.9% credibility score and 16 stars reflect early-stage work, but the core concept is sound and the multi-renderer integration is well thought out. If you're building AI presentation workflows, this is worth a look for the outline planning layer alone, though you should expect to iterate as the project matures.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.