MA

mbreakduga/ai-ugc-factory

Developer tools
20 stars Quality 54 Trend 14MIT

Open-source AI ad asset system and AI UGC batch production factory

Overview

中文 README · AIVideoPro.io · English website · 中文网站 · Demo Release is an open-source workflow for building enterprise-grade AI ad asset systems and producing AI UGC ad videos in batches. It is designed for founders, growth teams, agencies, and creators who want a reusable production line instead of one-off prompts: product intake, asset mapping, reference control, UGC script design, generation prompt packs, QA gates, demo packaging, and performance-learning loops. Matrix launch note: the full Matrix product link is coming soon. For early access or collaboration, see assets/contact/wechat-qr.jpg. - 🤖 A reusable that turns product assets into production-ready ad video briefs. - 🗂️ for real product images, webpages, references, videos, and contact assets. - ✍️ for UGC videos, demo reels, and ad-asset systems. - ✅ for voice, pacing, visuals, and product claims before delivery. - 🌐 with full video players. - 🎬 hosted as GitHub Release assets.

README

AI UGC Factory

中文 README · AIVideoPro.io · English website · 中文网站 · Demo Release

AI UGC Factory is an open-source workflow for building enterprise-grade AI ad asset systems and producing AI UGC ad videos in batches.

It is designed for founders, growth teams, agencies, and creators who want a reusable production line instead of one-off prompts: product intake, asset mapping, reference control, UGC script design, generation prompt packs, QA gates, demo packaging, and performance-learning loops.

Matrix launch note: the full Matrix product link is coming soon. For early access or collaboration, see assets/contact/wechat-qr.jpg.

What this project gives you

  • 🤖 A reusable AI UGC Factory skill that turns product assets into production-ready ad video briefs.
  • 🗂️ Asset manifest templates for real product images, webpages, references, videos, and contact assets.
  • ✍️ Prompt-pack examples for UGC videos, demo reels, and ad-asset systems.
  • ✅ QA gates for voice, pacing, visuals, and product claims before delivery.
  • 🌐 Separate English and Chinese GitHub Pages with full video players.
  • 🎬 Directly playable demo videos hosted as GitHub Release assets.
  • 🔁 A performance-learning loop that turns winning angles into reusable batch templates.

Demo videos

GitHub README filters embedded MP4 players, so the previews below are animated GIFs that play directly on the repository page. Click any preview to open the full MP4.

Why it exists

Most AI ad workflows fail because they treat video generation as a prompt-writing problem. Production quality usually comes from a system:

  1. real product assets are organized and named by function;
  2. reference images/videos are attached as model-readable inputs, not merely described;
  3. every 1-2 second beat has a job: hook, proof, benefit, objection, CTA;
  4. voice, pacing, visuals, and claims are QA’d before delivery;
  5. winning angles become reusable batches.

This repo packages that system into a compact open-source starter.

See it in motion — a 15s AI-generated explainer of the production line (GitHub strips inline video players, so the preview below is an animated GIF; click it for the full MP4):

Play full explainer MP4 · Watch on the English site

Repository structure

skills/ai-ugc-factory/        # reusable Matrix/agent skill
examples/briefs/              # product brief examples
examples/asset-manifests/     # asset inventory examples
examples/prompt-packs/        # prompt-pack examples
templates/                    # reusable production templates
docs/                         # GitHub Pages project website
assets/contact/               # early-access contact asset
scripts/                      # helper scripts

Quick start

  1. Copy templates/product-brief.md and fill in the product, audience, claims, and offer boundaries.
  2. Put real assets in a stable folder and build an asset manifest from templates/asset-manifest.csv.
  3. Use skills/ai-ugc-factory/SKILL.md as the production workflow.
  4. Draft a prompt pack with templates/video-prompt-pack.md.
  5. Run the QA checklist before rendering or publishing.

Contact / early access

Matrix launch link is coming soon. If you want early access, collaboration, or batch custom AI UGC production support, use the contacts below.

License

MIT. Use it, fork it, adapt it — just keep claims truthful and respect product assets you do not own.

View the full README on GitHub

Recommended Tools

Try a different keyword or remove a filter.

Install

npx skillfish add mbreakduga/ai-ugc-factory
New to MCP? Start with the mcpskill.org guide →