ad-copy-writer
Skill1 starsChanged 6 days agoUnlicenseView on GitHub
Writes advertising copy — taglines, headlines, body copy, CTAs, and full ad sets — with a structured brief intake, audience targeting, variant generation, and self-critique loop. Accepts campaign platforms from creative-director or raw briefs. Use when asked to write ad copy, taglines, slogans, headlines, product descriptions, campaign copy, launch copy, brand rallying cries, or go-to-market messaging. Do not use for UX microcopy (use ux-copy), full campaign ideation (use creative-director), media planning, or brand identity.
2 other projects use a version of this.
Adopted by
Most recently taken up by
- cognyai/claude-code-marketing-skillstheir own version, first seen Oct 12, 2026
- the-baweja/ad-copy-writertheir own version, first seen May 21, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- cognyai/claude-code-marketing-skills★ 106same name
- the-baweja/ad-copy-writer★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository76 other files it gives its agents.
- AGENTS.mdAGENTS.md
- ae-inspect-describeSkill
- algorithmic-artSkill
- animation-directorSkill
- assembly-guideSkill
- brainstorming-and-ideatingSkill
- brainstorming-research-ideasSkill
- brainstormingSkill
- brief-to-tasksSkill
- canvas-designSkill
- cli-designSkill
- cli-output-hierarchySkill
- commit-messageSkill
- craft-uiSkill
- creative-directorSkill
- design-briefSkill
- design-flowSkill
- design-mock-auditSkill
- design-reviewSkill
- design-system-patternsSkill
- design-tokensSkill
- deslopifySkill
- diagnoseSkill
- diagramming-app-routesSkill
- doc-restructureSkill
- eval-criticSkill
- formatting-open-markdownsSkill
- frontend-designSkill
- frontend-slidesSkill
- gdocs-tab-writerSkill
- gemini-apiSkill
- gif-creationSkill
- github_repo_researchSkill
- grill-meSkill
- grill-with-docsSkill
- hallmarkSkill
- handoffSkill
- html-presentationSkill
- humanizerSkill
- i-have-adhdSkill
- improve-codebase-architectureSkill
- information-architectureSkill
- install-skill-from-githubSkill
- markdown-output-hierarchySkill
- marketing-psychologySkill
- merge-latestSkill
- modern-web-performanceSkill
- motion-designSkill
- pdf-readerSkill
- pm-brainstormingSkill
- pm-discovery-brainstormSkill
- product-marketingSkill
- prototypeSkill
- pull-requestSkill
- ralph_loopSkill
- react-best-practicesSkill
- review-loopSkill
- sandbox-explainerSkill
- skill-creatorSkill
- skill-optSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 7 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈7
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
- Distinctive: line 15
Write advertising copy at the level of Wieden+Kennedy / Droga5 / Ogilvy.
- Distinctive: line 41
If a campaign platform, Big Idea, or insight already exists (e.g., from creative-director output), skip to Phase 2 or 3 — do not re-derive …
- Distinctive: line 62
If data is insufficient, ask 3-5 precise questions.
- Distinctive: line 119
One idea per sentence: If it has a comma, it probably needs a period.
- Distinctive: line 196
Always include: Rationale (why this works), What was killed (rejected directions and why), Tone calibration note (where this sits on the …
- Distinctive: line 238
From creative-director: If a campaign platform, Big Idea, or insight exists, use it as the foundation.
- Distinctive: line 241
From ux-copy: If writing in-product copy that also serves a marketing function (e.g., onboarding, upgrade prompts), coordinate tone with …
≈ “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.
Files
Ad Copy Writer
Write advertising copy at the level of Wieden+Kennedy / Droga5 / Ogilvy. Core principle: Simplicity as Violence — if it can't be understood in 3 seconds, rewrite it.
Style DNA: bold, direct, confident, occasionally funny. Never cute, never sentimental, never corporate-speak. The copy should feel like a smart friend who happens to know everything about the product — not a brochure.
Creativity = clarity + surprise. Generic but clear = forgettable. Surprising but confusing = wasted. Find the intersection of instantly understood and unexpectedly phrased.
Instructions
Phase Router
Determine the phase from context:
- New request / "write copy for" / raw brief → Phase 1: BRIEF
- Have a brief, need audience sharpening → Phase 2: AUDIENCE
- Have audience + brief, need copy → Phase 3: GENERATE
- "Improve this" / "make it punchier" / existing copy to review → Phase 4: CRITIQUE
- "Finalize" / "give me the set" / ready for delivery → Phase 5: DELIVER
- Full cycle (standard request) → sequentially Phase 1 → 2 → 3 → 4 → 5
If a campaign platform, Big Idea, or insight already exists (e.g., from creative-director output), skip to Phase 2 or 3 — do not re-derive strategy.
Phase 1: BRIEF (intake)
Extract from incoming material:
- Product/service: What is it? What does it do?
- Key benefit: The single most compelling thing (not a feature list)
- Target audience: Who, demographics, psychographics, pain points
- Platform/format: Where does this copy live? (billboard, social, Google Ads, landing page, email subject line, etc.)
- Tone constraints: Brand voice guidelines, mandatory language, legal disclaimers
- Competitive landscape: Who else is in this space? What do they sound like?
- Deliverable type: Tagline only / headline + body / full ad set / multi-platform variants
If data is insufficient, ask 3-5 precise questions. Not "tell me about the audience" — instead: "Who is the decision-maker? What's their biggest frustration with existing alternatives?"
Phase 2: AUDIENCE (targeting)
Sharpen the audience lens:
- Mindset, not demographics: "28-35 urban professionals" is a spreadsheet row. "People who feel guilty about not cooking but won't admit it" is an audience. Find the mindset.
- The language they actually use: What words does this audience use to describe their problem? Mirror that language.
- The tension: What does this audience want vs. what stands in their way? (Borrow from creative-director's tension-spotting if available.)
- The context of encounter: Where are they when they see this copy? Doom- scrolling at 11pm is a different headspace than commuting at 7am.
Output: one-sentence audience portrait in the format: "[Audience] who feels [tension] when [context]."
Phase 3: GENERATE (copy creation)
Read: references/copy-frameworks.md
Algorithm:
Select 2-3 frameworks from
references/copy-frameworks.mdbased on the brief type:- Brand awareness → Disruption, Provocation
- Direct response → PAS, AIDA
- Product launch → BAB, Feature-Benefit Flip
- Emotional resonance → Story Spine, Before/After
- Generate 8-12 variants across the selected frameworks
- Mark the first 3 as "warmup" (serial order effect — later variants are statistically more original). Don't delete them, but bias toward variants 5-12+
Each variant must include:
- Headline (if applicable)
- Body copy (if applicable)
- CTA (if applicable)
- Framework used (which approach generated this)
- For multi-platform requests, generate platform-native variants — don't just
truncate. Read
references/platform-constraints.mdfor format limits.
Copy Rules:
- One idea per sentence. If it has a comma, it probably needs a period.
- Verbs > adjectives. "Cuts your commute in half" beats "An amazingly fast solution."
- Specific > vague. "47 minutes" beats "less time." "$3/month" beats "affordable."
- Contrast creates memorability. "Tastes like Saturday. Priced like Tuesday." Structure: [desirable thing] + [unexpected qualifier].
- Cultural references land harder than generic claims. But only if the audience recognizes them.
Phase 4: CRITIQUE (self-assessment loop)
Read: references/style-calibration.md
PASS 1: Five-test evaluation
Apply each test to every variant. Score 1-10 per test.
| Test | What it measures | Fail signal |
| Billboard | Understandable in 3 seconds at | Requires re-reading |
| : : speed? : : | ||
| Swap | Replace brand with competitor — | Generic if yes |
| : : still works? : : | ||
| Bar | Would you repeat this to a friend at | Too corporate / stiff |
| : : a bar? : : | ||
| Delete | Remove any word — does meaning | Flab if no |
| : : change? : : | ||
| Overnight | Will you remember this tomorrow? | Forgettable |
Weighted score: Billboard (0.25) + Swap (0.25) + Bar (0.20) + Delete (0.15) + Overnight (0.15)
Select top 3.
For each top 3, answer: "Why isn't this a 9?"
PASS 2: Targeted improvement (if top < 8.5)
For each of the top 3:
- Identify the weakest test
- Apply a DIFFERENT framework from
references/copy-frameworks.md - Rewrite with targeted fix
- Re-score
PASS 3: Final or restart
- Score >= 8.5 → EXIT → Phase 5
- Score 7.0-8.4 and improving → one more pass with new framework
- Score < 7.0 OR plateau (delta < 0.3) → RESTART from Phase 3 with different framework set
- Maximum 4 passes total. If 4 passes completed, deliver best with honest assessment
Stopping Criteria
(a) Top variant >= 8.5 → exit with final deliverable (b) 4 passes completed → deliver best with honest note (c) Two consecutive passes with delta < 0.2 → plateau, deliver with note
Phase 5: DELIVER (final output)
Read: assets/output-templates.md
Format depends on the request type:
- Tagline only → 3 options ranked with rationale
- Headline + body → Top recommendation + 2 alternatives
- Full ad set → Headline / subhead / body / CTA as a unit
- Multi-platform → Platform-native variants in a comparison table
- Campaign copy → Unified voice across 3-5 executions
Always include: Rationale (why this works), What was killed (rejected directions and why), Tone calibration note (where this sits on the spectrum).
Anti-Patterns (with examples)
The Pun Reflex
- NO: "Lettuce help you eat better! 🥬"
- YES: "You know what you should eat. We made it easy."
The Feature List Disguised as Copy
- NO: "With 128GB storage, 5G connectivity, and an A17 chip..."
- YES: "Every photo you've ever taken. In your pocket. Always."
Corporate Warm Porridge
- NO: "We're committed to delivering innovative solutions that empower..."
- YES: "It works. First try. Every time."
The Cute Tax
- NO: "Your wallet will thank you 🙏✨"
- YES: "Half the price. Same everything."
The Vague Aspiration
- NO: "Reimagine what's possible."
- YES: "Ship on Monday. Not 'sometime in Q3.'"
The Thesaurus Crime
- NO: "An unparalleled, best-in-class, paradigm-shifting experience"
- YES: "Better than anything else. We checked."
Interop
- From creative-director: If a campaign platform, Big Idea, or insight exists, use it as the foundation. Don't re-derive strategy — write copy that executes the existing strategic direction.
- From ux-copy: If writing in-product copy that also serves a marketing function (e.g., onboarding, upgrade prompts), coordinate tone with ux-copy principles but maintain ad-copy directness.
References
references/copy-frameworks.md— AIDA, PAS, BAB, and 10+ more as actionable cardsreferences/style-calibration.md— detailed rubric for each score with do/don't examplesreferences/platform-constraints.md— character limits and format rules per platformassets/output-templates.md— output format templates by deliverable type
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
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.

