agentleFS
Sign inStart free
fourteenwmfourteenwm/ppc-ai-skills/ad-copy-generation-framework/SKILL.md

ad-copy-generation-framework

Skill72 starsChanged 4 months agoMITView on GitHub

Comprehensive 23-element Google Ads RSA copywriting framework with distribution formulas, sentiment scoring, and the "Four Words to Value" rule. Auto-invoke when user says "generate RSA copy", "write ad copy", "RSA framework", "headlines and descriptions", or asks for ad copywriting principles. Pure protocol skill, no scripts required.

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 6 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Tier 1 (Mandatory): Always use - keyword placement, funnel matching, title case, CTA, continuity

    line 40
  2. Distinctive:

    Tier 3 (Optional): Use strategically - FOMO, puns, competitor comparisons, DKI

    line 42
  3. Distinctive:

    Use symbols/numbers where possible (!, ?, $, %)

    line 67
  4. Distinctive:

    Start with verb (action-oriented)

    line 75
  5. Distinctive:

    Use most important keyword

    line 77
  6. Distinctive:

    Avoid "we" - use "you" (customer-focused)

    line 78

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Ad Copy Generation Framework

Comprehensive Google Ads RSA copywriting framework with 23 elements, distribution formulas, and "Four Words to Value" rule for creating high-performing ad copy.

Auto-Invoke When

  • Generating RSA headlines or descriptions
  • Writing ad copy for Google Ads
  • Optimizing existing ad copy
  • Reviewing ad copy quality
  • Training on ad copywriting principles

What This Skill Deliberately Does NOT Do

  • Verify claims. The framework governs structure and distribution — how many keyword headlines, where the CTA goes, what makes copy strong. Whether a specific claim may be used at all is ad-copy-verification-standard's call: every claim must come from explicitly verified website content, and no distribution formula overrides that.
  • Write to live accounts. Output is copy for human review — a Google Sheet reviewed before import, never a direct API write.
  • Decide which ads need new copy. Audit and refresh skills pick the targets; this framework loads once the writing starts.
  • Check Google Ads policy. Character limits, title case, and sentiment targets (technical-specs.md) are formatting validation, not a policy or trademark review.

Key Features

  • 3-Tier Element System: Mandatory, Preferred, and Optional elements
  • Distribution Formula: 3 keyword + 2 social proof + 4 generic USP + 2 CTA + 1 pun + 3 flexible
  • "Four Words to Value" Rule: Website headline optimization for cognitive momentum
  • Character Constraints: 30 char headlines, 90 char descriptions (Google Ads limits)
  • Sentiment Scoring: Target 0.8-1.0 positive sentiment
  • Type Categorization: Headlines tagged by strategic type (keyword-focused, benefit-driven, emotional, etc.)

Framework Components

1. The 23 Elements

Organized into three tiers:

  • Tier 1 (Mandatory): Always use - keyword placement, funnel matching, title case, CTA, continuity
  • Tier 2 (Preferred): Strongly recommended - punctuation, timely language, features+benefits, single focus
  • Tier 3 (Optional): Use strategically - FOMO, puns, competitor comparisons, DKI

See framework.md for complete list.

2. Distribution Formulas

Headlines (15 total):

  • 3 keyword-focused
  • 2 social proof
  • 4 generic USPs
  • 2 CTAs
  • 1 pun
  • 3 flexible

Descriptions (4 total):

  • 2 with keyword + product USP + CTA
  • 2 with keyword + generic USPs (no CTA)

See distribution.md for details.

3. Technical Specifications

  • Headlines: 30 characters max (hard Google Ads limit)
  • Descriptions: 90 characters max (hard Google Ads limit)
  • Capitalize Each Word (Title Case)
  • Use symbols/numbers where possible (!, ?, $, %)
  • Positive sentiment (0.8-1.0 target)

See technical-specs.md for validation rules.

4. "Four Words to Value" Rule

Website headlines should deliver value in first 4 words:

  • Start with verb (action-oriented)
  • Subject/predicate structure (cognitive momentum)
  • Use most important keyword
  • Avoid "we" - use "you" (customer-focused)
  • Imply value, not work

Example: "Get Emergency Plumbing Fixed Today" ^ ^ ^ ^ ^ Verb You Benefit Action Timeframe

See examples.md for more.

When to Load Other Skills

SkillWhen
ad-copy-verification-standardBEFORE generating — it gates what claims are allowed in (explicitly verified website content only); this framework then shapes those claims into headlines and descriptions
mutation-safetyOnly if generated copy is pushed to an account via API — the normal output path (Google Sheet, manual review before import) is not a mutation

Producing skills in this repo that route here for copywriting: rsa-single-account, rsa-refresh, rsa-bulk-edit, pmax-builder, and competitor-analysis-v2.

Usage in Agents

Agents should invoke this skill when generating ad copy. The skill provides:

  • Complete 23-element framework reference
  • Distribution formulas for headline/description balance
  • Validation rules for character limits
  • Examples of good vs bad ad copy

Typical agent workflow:

  1. Invoke skill to get framework
  2. Gather inputs (keyword, USPs, reviews, business overview)
  3. Generate headlines/descriptions following distribution formula
  4. Validate against technical specs
  5. Output with character counts and types

Files in This Skill

  • SKILL.md - This file (auto-invoke description)
  • framework.md - Complete 23-element framework with principles
  • distribution.md - Headline/description generation formulas
  • technical-specs.md - Character limits, formatting, validation rules
  • examples.md - Good/bad examples, "Four Words to Value" demos

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free