| name | AI Carousel Content Engine |
| description | Turn articles, raw ideas, transcripts, or brand insights into 7-10 slide carousel packages with copy, visual direction, GPT Image 2 prompts, review checklists, and publish packs. |
| version | 1.0 |
| format | completion-engine |
| expert | AI Carousel Content Engine |
| domain | social carousels, GPT Image 2 prompts, content repurposing, brand style systems, marketing asset production |
| workflows | 7 |
| routing | long-tail |
AI Carousel Content Engine
This skill creates Farrice's own version of the article-to-carousel workflow demonstrated in the Luke Carter video. It does not copy the hidden community prompt. It builds a reusable capability from observed workflow mechanics, existing Antigravity carousel frameworks, GPT Image 2 prompt direction, and human-in-the-loop creative review.
Core Promise
Turn a source idea into a visual marketing asset fast:
- Start with an article, idea, transcript, or client insight.
- Extract the strongest claims and sequence them into a 7-10 slide carousel.
- Generate a style board and GPT Image 2 structured prompt.
- Review for taste, text accuracy, brand fit, and strategic CTA.
- Package for Instagram, LinkedIn, client delivery, or a weekly content pipeline.
Available Workflows
Local CLI
python3 execution/ai_carousel_engine.py --mode pack --source path/to/article.md --title "Topic"
python3 execution/ai_carousel_engine.py --mode copy --text "raw idea"
python3 execution/ai_carousel_engine.py --mode prompt --source path/to/carousel.md --style "brand/style direction"
Output Package
Each full run writes to deliverables/ai-carousel-engine/<slug>/:
source.md
carousel-script.md
slide-brief.json
gpt-image-2-prompt.json
style-board.md
review-checklist.md
publish-pack.md
Stack With
| Workflow | Role |
|---|
| GPT Image 2 (direct) | Use gpt-image-2-prompt.json directly as a GPT Image 2 layout prompt; any actual image generation must flow through the cost-gated visual route (creative_router.py pre-flight, e.g. skills/fantastic-posters/) |
/mood-board | Build visual references before prompt generation |
/design-md-synthesize | Create a durable brand/design system before recurring carousels |
/creative-review | Taste-check generated outputs |
/watch (execution/fetch-video-context.py) | Extract source material from videos before carousel creation |
/extract-forge | Build a deeper skill if a carousel system should become a broader content engine |
Quick Reference
Execution Prompts (structure-pure v2)
6 deterministic practitioner prompts — each carries an Output Contract, Output Skeleton, and Quality Gate. When a deliverable matches one, Read it and honor its contract instead of improvising the output shape.
- Carousel Script: [TITLE] —
skills/ai-carousel-content-engine/references/prompts-v2/carousel-copy-script.md
- Carousel Review — [TITLE] —
skills/ai-carousel-content-engine/references/prompts-v2/carousel-review-repair.md
- [TITLE] — Full Carousel Package —
skills/ai-carousel-content-engine/references/prompts-v2/full-carousel-package.md
- AI Carousel Content Engine — GPT Image 2 Design Prompt —
skills/ai-carousel-content-engine/references/prompts-v2/gpt-image-2-design-prompt.md
- Publish Pack — [TITLE] —
skills/ai-carousel-content-engine/references/prompts-v2/publish-pack.md
- Style Board — [BRAND_OR_CLIENT] —
skills/ai-carousel-content-engine/references/prompts-v2/style-board.md