Advertising Skills for AI Agents

The direct-response, paid media, and copywriting system for serious operators.

Before vs After

Before:
“Scale your business with proven strategies”

After:
“You’re not bad at ads.
You’ve just never been shown how to structure them properly.”

Same business.
Completely different outcome.

Most AI marketing outputs sound the same.

Generic.
Polished.
Forgettable.

This repository fixes that.

It encodes the thinking behind:

• high-performing ads
• high-converting funnels
• sharp positioning
• and direct-response copy that actually sells

Not as prompts.

As skills.

What This Is

A best-in-class skill library for AI agents built around:

• Paid acquisition (Meta, TikTok, YouTube)
• Eugene Schwartz-style copywriting
• Offer design and positioning
• Funnel architecture and conversion logic
• Creative testing and scaling
• Campaign diagnosis and optimisation

Each skill is:

• focused
• opinionated
• reusable
• and designed to work together

Why This Is Different

Most “AI marketing libraries” optimise for output.

This optimises for thinking.

That means:

• It doesn’t guess the audience — it defines it
• It doesn’t write copy first — it builds the strategy first
• It doesn’t try to sound smart — it tries to convert
• It doesn’t just create — it filters, qualifies, and diagnoses

It combines two worlds that are usually separate:

The Copy Chief Brain

• Awareness levels (Eugene Schwartz)
• Market sophistication
• Mechanisms and big ideas
• Headlines, leads, proof, objections, close

The Operator Brain

• Ad creative and angles
• Funnel paths and conversion logic
• Qualification and filtering
• Testing and performance diagnosis

Most systems have one.

This has both.

How It Works

Each skill lives in its own folder:
skills///SKILL.md

Each skill defines:

• when to use it
• what inputs it needs
• how it thinks
• what it outputs
• what it should not do

They are designed to be chained.

Example Flow

Building a booked-call campaign:
avatar-extraction
→ offer-extraction
→ schwartz-awareness-mapper
→ mechanism-builder
→ ad-angle-multiplier
→ scroll-stopping-creative
→ conversion-path-builder
→ objection-crusher
→ generic-language-killer

This produces:

• clear positioning
• strong angles
• high-performing ads
• a clean funnel
• and conversion-ready messaging

Example Output

See:
/examples/booked-call-campaign-ai-ads.md

Core Skill Categories

Foundations

Define reality before writing anything.

• avatar-extraction
• voice-of-customer-miner
• offer-extraction
• proof-harvester
• positioning-core

Copy Chief

The Eugene Schwartz layer.

• schwartz-awareness-mapper
• market-sophistication-analyzer
• dominant-desire-extractor
• mechanism-builder
• headline-matrix
• objection-crusher
• close-architect

Operator OS

Execution and scale.

• scroll-stopping-creative
• ad-angle-multiplier
• conversion-path-builder
• landing-page-architect
• email-sequence-architect
• performance-diagnosis
• testing-roadmap

QA

What makes the output actually good.

• generic-language-killer
• claim-checker
• consistency-auditor
• final-polish

Orchestrators

Pre-built multi-skill flows.

• full-funnel-campaign-orchestrator
• booked-call-campaign-orchestrator
• campaign-diagnosis-orchestrator

Who This Is For

• Media buyers who want better creative and messaging
• Founders running paid ads who are tired of guesswork
• Agencies who want consistency across teams
• Operators building AI-driven marketing systems

What This Is Not

• Not generic content marketing templates
• Not “write me 10 captions” prompts
• Not surface-level frameworks

This is built for:

• direct response
• paid traffic
• real budgets
• real consequences

Installation

Compatible with any agent using the Agent Skills pattern.

Install into:
.agents/skills/

Works with:

• Claude Code
• Cursor
• Windsurf
• Codex
• custom agent frameworks
• Open Claw

Philosophy

The future isn’t better prompts.

It’s better encoded thinking.

Roadmap

• Expanded Copy Chief library (mechanisms, big ideas, proof systems)
• Advanced funnel orchestration
• Creative testing systems
• Platform-specific skill layers (Meta, TikTok, YouTube)

License

MIT

Author

Kim Barrett
Advertising systems, AI-driven marketing, and direct-response strategy

If you use this properly, it won’t just improve your ads.

It will change how you think about them.