Product Marketing Asset Generator

Problem

Product marketing materials are often scattered across landing pages, UI screenshots, demo videos, and PRD text. Teams need publishable assets—posters, short-video scripts, and user manuals—rather than repeatedly rewriting the same product facts. The hard part is keeping copy, visuals, and manuals consistent while producing prompts that can be used directly by image-generation tools.

How It Works

The skill parses input into a Product Representation Object (PRO), then runs a narrative engine that generates poster, video, and manual assets in parallel. Poster output includes headline, subheadline, layout type, generation-ready visual prompts, and 2–3 variants. Video output structures a 30-second script into Hook, Problem, Solution, Feature, and CTA, with each scene specifying visual, voiceover, and screen text. Manual output covers Quick Start, step-by-step guidance, FAQ, and troubleshooting. The pipeline requires structured JSON, avoids hallucination, and maintains consistency across assets.

Boundaries

It is best suited to inputs with clear product facts and is not intended to invent benefits. If URL, image, or video parsing fails, it falls back to text extraction, OCR/vision, or keyframe extraction. The output emphasizes editable marketing assets and prompts; final rendering still depends on downstream image, video, or documentation tools.