visual-presentation-production

Generate 16:9 image-based slide decks through gated brief, outline, and style approvals, delivered as one PDF.

Updated Aug 22, 2026
One-click install
npx skills add https://github.com/JustinChangTW/ai-skills-core --skill visual-presentation-production-justinchangtw
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: visual-presentation-production
Source: https://github.com/JustinChangTW/ai-skills-core/tree/main/skills/07-presentations-visual/visual-presentation-production
Command: npx skills add https://github.com/JustinChangTW/ai-skills-core --skill visual-presentation-production-justinchangtw

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires Pillow, and includes scripts (resource) and references (resource) components.

What problem does it solve? Creating consistent, on-brand image-based presentations often fails because slides are generated without an approved brief, a fixed page count, or a locked visual style, resulting in mismatched layouts and unreliable text. This Skill enforces mandatory approval gates so every deck is planned, styled, and quality-checked before delivery. ## Core Features & Use Cases - Gated production workflow: Requires an approved brief, exact slide-count outline with per-page layout descriptions, a locked visual style profile, and an approved sample slide before any generation begins. - Per-slide image generation with QA: Generates one self-contained 1920x1080 PNG per slide using the host's native image capability, then validates each page against its outline and the global style profile. - PDF-only delivery with validation script: The bundled Python script validates the outline, manifest, and slide images, then assembles everything into a single presentation.pdf with a QA report. - Use Case: A workshop instructor needs a 6-page branded lesson deck. The Skill confirms the brief and page count, locks a style preset, generates and QA-checks each slide, and delivers one print-ready PDF. ## Quick Start Use visual-presentation-production to create a 6-page 16:9 presentation about my workshop topic, confirming the brief, outline, and style with me before generating each slide and delivering one PDF.

Frequently Asked Questions about visual-presentation-production

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create an image-based presentation deck with AI?▼

Start by approving a brief covering purpose, audience, exact slide count, and visual direction. The Skill then produces an outline with per-page layouts, locks a style profile, generates one 1920x1080 image per slide, and delivers a single PDF after QA.

How to generate slides with a consistent visual style?▼

Lock a visual style profile before generating any slide, either from uploaded brand assets or a preset like Editorial or Academic Clean. One representative sample slide must be approved first, and every subsequent page is checked against that profile during QA.

Can I get a PPTX or Google Slides file instead of PDF?▼

No, the only user-facing delivery is presentation.pdf. The Skill explicitly does not offer PPTX, Google Slides, speaker notes, or PNG bundles; approved slide images are internal inputs only.

Does this work in ChatGPT without running scripts?▼

Yes, in ChatGPT Work it uses the host's native image generation and PDF creation capabilities only. The bundled Python assembly script runs only in local environments like Codex with Pillow installed.

Why does slide generation stop before creating any images?▼

Generation is blocked until every gate is approved: complete brief, exact page-count outline with layouts, locked style profile, and an approved sample slide. Missing brief fields or unconfirmed source material marked as pending will halt the workflow.