ad-library-research
Skill0 starsChanged 21 days agoMITView on GitHub
Research any brand's live ads for free with the public ad libraries (Meta, Google, TikTok, LinkedIn, Snap) and turn what you find into hooks and angles.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- xhoantran/scrapeio-agenttheir own version, first seen Nov 8, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- xhoantran/scrapeio-agent★ 66same 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 repository6 other files it gives its agents.
- ad-copy-frameworksSkill
- ad-hooksSkill
- ad-teardownSkill
- campaign-analystSkill
- social-media-specsSkill
- ugc-ad-scriptsSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
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 9
Every major platform publishes the ads running on it as an official transparency tool, for exactly this kind of research.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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-library research (no account needed)
Every major platform publishes the ads running on it as an official transparency tool, for exactly this kind of research. This skill is the map: where each library is, what it can filter, what it hides, and how to turn a pile of ads into something a marketer can use.
Where to look
| Library | URL pattern | What it shows | Limits |
| Meta Ad Library (Facebook, Instagram, Messenger, Threads) | https://www.facebook.com/ads/library/?active_status=active&ad_type=all&country=US&q=<brand or keyword>&search_type=keyword_unordered | creative, copy, start date, platforms; EU results add reach and targeting | no spend for non-political ads; "active" vs "all" matters, because the longest-running ads are the proven winners |
| Google Ads Transparency Center | https://adstransparency.google.com/?region=US&domain=<brand.com> | Search, Display, YouTube creatives by advertiser or domain, with first/last-shown dates | no copy for many Search ads unless you open each; filter by format |
| TikTok Creative Center (Top Ads) | https://ads.tiktok.com/business/creativecenter/inspiration/topads/pc/en | top-performing TikTok ads by region, industry, objective, with likes and CTR proxies | curated sample, not the whole library |
| TikTok Ad Library (EU transparency) | https://library.tiktok.com/ads?region=all | every ad shown in the EU, searchable by the advertiser's LEGAL entity name | search is exact on the registered entity ("Duolingo, Inc."), EU only |
| LinkedIn Ad Library | https://www.linkedin.com/ad-library/search?companyName=<Company> | B2B ads by company or keyword, with impressions by country | needs a signed-in LinkedIn session in the browser |
| Snap political/issue ads | https://www.snap.com/en-US/political-ads | political and issue ads with spend | political only |
Use these the way a person does
Open the pages in a normal browser session, one query at a time, and read what they show. These are transparency tools the platforms publish for the public; their terms prohibit scripted bulk collection, scraping, or working around rate limits, so do not automate requests against them from the agent. For automated, licensed research use a tool built for it.
Procedure
- Start with the brand's own ads, then 3 to 5 direct competitors, then 2 brands outside the category that are famous for great creative (that is where the copyable ideas usually are).
- Sort by longevity, not recency. An ad still running after 60+ days is paying for itself. In Meta's library use "active" first, then "all" to see what they retired.
- Log each ad in one line: brand · format (image/video/carousel) · hook mechanic · angle · offer · run length.
- Cluster the angles. Five ads about "no more back pain" and one about "fits under a suit" tell you which angle the market has already validated and which is under-served.
- Write the brief from the clusters: the top 3 angles, the hook mechanic each uses, and one specific thing to do differently. Quote the actual headlines; paraphrase loses the mechanic.
Reading a creative fast
- First 2 seconds: what stops the thumb (see the
ad-hooksskill for the mechanics by name). - Is the product on screen inside 3 seconds? Long-running DTC video almost always says yes.
- Who is talking: brand voice, creator, or nobody (product-only). Note the ratio across the winners.
- Ending: offer + CTA, or a soft "learn more"? Match the strongest cluster.
Next
Tear the strongest brand down properly with the ad-teardown skill, and write your own versions with ad-hooks
and ugc-ad-scripts.
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.

