ad-copy-generator
Skill5 starsChanged 17 days agoMITView on GitHub
Writes high-converting ad creative and copy variations for Meta (Facebook and Instagram), TikTok, Google Search, YouTube, and LinkedIn ads using direct-response frameworks, respecting each platform's character limits and policies, and organizes them into an A/B test matrix. Use when the user asks for ad copy, ad creatives, ad scripts, headlines, or paid-social or PPC campaign text.
4 other projects use a version of this.
Adopted by
Most recently taken up by
- milad-ahmd/ad-copy-generatorword for word, first seen Nov 8, 2026
- TheMattBerman/meta-ads-kittheir own version, first seen Oct 8, 2026
- marcosmodly/marketing-skilltheir own version, first seen Oct 8, 2026
- whyashthakker/agent-skills-marketingtheir own version, first seen Jun 21, 2026
Other versions4 projects use a version of this.
Word for word · 1
- milad-ahmd/ad-copy-generator★ 0100% same
Edited · 0
None found.
Wrote their own version · 3
- TheMattBerman/meta-ads-kit★ 331same name
- marcosmodly/marketing-skill★ 100same name
- whyashthakker/agent-skills-marketing★ 15same 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 repository56 other files it gives its agents.
- docs/llms.txtllms.txt
- a11y-fixerSkill
- agent-guardrailsSkill
- agent-memory-keeperSkill
- ai-slop-detectorSkill
- api-to-sdkSkill
- app-store-optimizerSkill
- architecture-doctorSkill
- brand-voice-guideSkill
- bug-detectiveSkill
- changelog-from-gitSkill
- codebase-explainerSkill
- codebase-time-machineSkill
- commit-craftSkill
- competitor-spySkill
- content-repurposerSkill
- cost-cutterSkill
- demo-recorderSkill
- dependency-doctorSkill
- dockerize-anythingSkill
- durable-agent-workflowsSkill
- e2e-flow-testerSkill
- email-campaign-writerSkill
- env-secrets-guardSkill
- gh-actions-auditorSkill
- hook-generatorSkill
- incident-commanderSkill
- influencer-outreachSkill
- instagram-growth-managerSkill
- landing-page-roasterSkill
- launch-post-writerSkill
- meeting-notes-to-ticketsSkill
- merge-conflict-resolverSkill
- migration-wizardSkill
- plan-grillerSkill
- prompt-injection-scannerSkill
- pr-reviewer-proSkill
- readme-glow-upSkill
- roast-my-githubSkill
- screenshot-to-componentSkill
- seo-blog-writerSkill
- skill-creator-proSkill
- skill-finderSkill
- skills-doctorSkill
- social-analytics-reporterSkill
- social-media-calendarSkill
- social-media-managerSkill
- sql-optimizerSkill
- startup-idea-validatorSkill
- tdd-loopSkill
- terse-modeSkill
- test-gap-hunterSkill
- tiktok-viral-creatorSkill
- ui-tasteSkill
- youtube-script-writerSkill
- youtube-seo-optimizerSkill
The file
We split this skill into the rules it gives an agent
3 rules. 0 are common, 3 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 3 edited versions we read keep these.
3
Distinctive
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 36
Follow ad policies: no false or unprovable claims, no fake scarcity, no "before/after" images for health or weight, no personal-attribute …
- Distinctive: line 37
Use only the proof and numbers the user provides.
- Distinctive: line 38
Keep the message consistent with the landing page.
“Common” is measured: kept by at least 6 in 10 of the 3 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Ad Copy Generator
Ads that stop the scroll, speak to one person and drive one action.
1. Brief
Product or offer, price, target audience (personas), main pain and desire, proof (reviews, numbers, guarantees), objections, campaign goal (awareness, traffic, leads, sales), platforms, and the landing page URL.
2. Angles (pick 4–6, then write variations for each)
Pain-agitate-solve · Desired outcome · Social proof / testimonial · Us vs them · Objection crusher · Urgency or offer (only if real) · Founder story · "How it works" demo · UGC-style.
3. Platform specs
| Platform | Output per angle |
| Meta (FB/IG) | Primary text: hook in the first 125 characters, short and long version · Headline ≤ 40 characters · Description ≤ 30 characters · CTA button |
| TikTok / Reels ads | A 15–30 second UGC script: hook (0–3s) → problem → product demo → proof → CTA, with on-screen text and shot list |
| Google Search (RSA) | 15 headlines of ≤ 30 characters and 4 descriptions of ≤ 90 characters, with the keyword in some headlines and pin suggestions |
| YouTube | A 6s bumper, plus a 15–30s skippable script with the brand and hook in the first 5 seconds |
| Intro text of ≤ 150 characters to avoid truncation, a headline, and a B2B outcome plus proof |
Count the characters and show the count for every limited field.
4. Test matrix
A table of angle × hook × visual concept with a hypothesis and the primary metric (CTR, CPC, CPA, ROAS). Recommend starting with 3–5 creatives per ad set and changing only one variable per test.
Rules
- Follow ad policies: no false or unprovable claims, no fake scarcity, no "before/after" images for health or weight, no personal-attribute phrasing ("Are you overweight?"), and the required disclaimers for finance or health.
- Use only the proof and numbers the user provides.
- Keep the message consistent with the landing page.
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.

