Omni Ad — AI skill for Claude Code
Claude Code skill: turn any product into an AI ad (GPT Image 2.0 poster + Google Omni Flash video) with hallucination-proof text locks.
How to install Omni Ad
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open aipauloshimas/omni-ad and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Omni Ad does
Claude Code skill: turn any product into an AI ad (GPT Image 2.0 poster + Google Omni Flash video) with hallucination-proof text locks.
Alternatives in AI
- Generative Media Skills — Multi-modal Generative Media Skills for AI Agents (Claude Code, Cursor, Gemini CLI) 3k ★
- Adopt AI Properly — Workflow recipe — the org-side AI adoption arc, policy to proof, by chaining 4 skills 1.3k ★
- Wax — Shared Single-file memory layer for all your agents, sub mili-second RAG over text, photo and video on Apple S 791 ★
README
omni-ad
A [Claude Code](https://claude.com/claude-code) skill that turns any product into a complete AI ad — with prompts your image model can't hallucinate fake specs into.
You give it a product name (+ a product photo when you generate). It researches the product's real standout features on the web, then hands you two ready-to-paste prompts:
- Poster prompt (GPT Image 2.0) — a stylized 9:16 campaign poster where every word on the poster is locked to an "ONLY text allowed" list built from sourced, official claims.
- Video prompt (Google Omni Flash) — a frozen, calibrated editing prompt that animates the poster into a full ad. Fast-paced but never chaotic. The poster carries all the branding, so this prompt never changes per product.
How it works
you: /omni-ad WHOOP 5.0
↓
skill asks: with model or product-only?
↓
skill researches the product (official tech names, headline claims, tagline)
↓
you get: poster prompt + video prompt + sources, saved to a .md file
↓
generate poster (GPT Image 2.0, 9:16) → use it as reference (Google Omni Flash) → done
Example spec callouts it produced for real products — all sourced, zero invented:
HEALTHSPAN — WHOOP AGE & PACE OF AGING·HEART SCREENER — FDA-CLEARED ECG·14+ DAY BATTERY — ALWAYS ON(WHOOP 5.0)LIGHTSTRIKE PRO EVO — ULTRALIGHT FOAM·ENERGYRODS — CARBON ROD SYSTEM·UNDER 100G — RACE DAY READY(Adidas Adizero Adios Pro Evo 3)
Install
git clone https://github.com/aipauloshimas/omni-ad ~/.claude/skills/omni-ad
Windows (PowerShell):
git clone https://github.com/aipauloshimas/omni-ad "$env:USERPROFILE\.claude\skills\omni-ad"
That's it. Open Claude Code and run `/omni-ad `.
Prerequisites
- Claude Code with web search available (used to research real product specs).
- Access to GPT Image 2.0 (poster) and Google Omni Flash (video) on whatever generation platform you use — the skill only writ
Related Skills
Vox Animated Ad MCP
Vox-style paper-collage animated video skill for AI agents — Claude + MaxFusion AI MCP (Google Omni). One sent
AI Video Editor
AI video editor - turn raw footage into multiple professionally edited versions from prompts. Color grading, z
Visual Skills
AI film director skills for agents: cinematic dramaturgy (Murch, blocking, montage) + exact prompt syntax for
Vibe Sort
Proof-of-concept Ruby gem that sorts arrays using LLM APIs: OpenAI, Anthropic Claude, Google Gemini, Groq, and
Claude Skill Video Transcribe
视频转文字工具:有字幕抓字幕,没字幕 AI 听写。支持 YouTube/B站/本地文件,Gemini 2.5 Flash 转录引擎
Ecom Video Seedance Prompt
这是一个基于deepseek-v4-flash-vision 视觉deepseek大模型的复刻带货爆款视频Skill, 会生成复刻效果的seedance提示词! 你刷到一条带货爆款视频,想让 AI 照着再来一条——本 s
Related Agents
Thumbnail Agent
Designs three deliberately different thumbnail concepts per video — Safe, Curiosity and Wildcard — each shippi
Media Lead
Use for content — post copy, image briefs (text-to-image prompts), short video scripts, brand voice, content c
Thumb Designer
썸네일 디자이너. 리서치 결과를 바탕으로 썸네일 컨셉을 설계하고, OpenAI gpt-image-2 이미지 생성으로 실제 썸네일을 제작한다.