AdLibrarySpy

100% free Shopify store and Meta ads intelligence, built for the community.

Find the stores that are winning, see the ads they run, and track your competitors.
Every number comes from a real measurement with its source attached. Nothing is estimated.

Use it with your own AI. Paste this into Claude, ChatGPT, Cursor or whichever AI you use:

Read https://adlibraryspy.com/SKILL.md and follow it to research my competitors' Shopify stores and Meta ads.

Open the app → · Run it yourself · Use it from Claude or ChatGPT · This week's breakouts

What it is

AdLibrarySpy indexes 14.7M online stores, 3.9M of them on Shopify (as of September 2026), and the
Meta ads they run. It shows you which stores are growing, what they sell and how they advertise.
Traffic figures come from SimilarWeb's measurement of each store's exact domain, labelled with the month
they cover. Ad data comes from the Meta Ad Library.

It's free: there are no plans, no credit limits and no checkout. It's also open source under MIT, and
this repository is the code that runs adlibraryspy.com.

Features

Shops: find the stores that are winning

• Filter by platform (Shopify, WooCommerce, Shopline, Shoplazza and others), category, traffic,
  traffic growth, product count, store origin, visitor country, creation date, technology and pixels,
  plus language, currency, theme, apps and socials once a store's details are indexed
• Preset views for one-click lists, and rows that show the homepage screenshot, country flag, visitor
  countries, live-ads trend, targeted countries, latest ad thumbnails and store launch date
• Picked for you: stores suggested from the niches you open. Skip one and the next slides in
• SimilarWeb visits and month-over-month growth for the exact store, with a trend line
• AOV, live Meta ads, 7-day ad peak, top products and launch year on every row
• Hide stores you've already seen, and save stores to shared folders
• Export CSV: the current filters and sort, up to 1,000 rows, ready for Excel or Sheets

Ads: the creative library

• Search ad copy, brands and landing pages, then filter by creation date, media type, format, placement,
  country and niche
• VSL filter for long video sales letters (2+ minutes), and one-tap date presets (7, 30, 90 days, this year)
• Ads show online stores by default. Run dates and placements for each ad, with video playback in the grid
• AI creative labels: hook, angle, funnel stage, offer and urgency, each with the model's confidence
• Save ads to folders and share them with your team. The Advertisers screen ranks brands by creative count
• Export CSV of the current search: ad and page ids, run dates, status, format, placements, copy, landing and media URLs

Store dossier: everything about one store

• Monthly visits and traffic over time, plus visitor countries
• Live Meta ads over time, with a link to the store's ads
• Product count, best sellers and latest products, the theme, currency and locale
• The pixels and apps the store runs, and similar shops

Brandtracker: watch your competitors

• Track any store. A snapshot of its traffic, live ads and new ads is recorded every day
• See what changed across 1, 7, 14 and 30 days, and organise tracked stores in folders
• Daily alerts by email (or weekly, or off): new ads, live-ad and traffic moves for the brands you track, new
  results for the Shops and Ads searches you save, and the stores that added the most live Meta ads the day
  before, in your niche. Sent only when something changed

Trends and the Monday report

• Trends: what advertisers launched on Meta this week vs last — products getting the most new ads,
  stores launching the most ads, brand-new advertisers, niches and ad formats gaining share — plus the
  niches and stores whose measured traffic is breaking out this month
• The weekly report: stores scaling their ads, the fastest traffic growth, ad peaks and the newest
  winners, published every Monday at /weekly and sent by email if you opt in
• Growing with no ads: fast-growing stores with no Meta ads, so you can tell paid growth from organic
• Landing pages (/landing-pages): every non-product page that
  ads send traffic to, ranked by how many ads point at it
• Public, shareable pages for every store (/store/{domain}), a store directory by niche, country and
  tech, and /trending

Ask your AI

A built-in MCP server (OAuth 2.1) lets Claude, ChatGPT and other assistants search shops and ads,
open dossiers and manage your brandtracker with your workspace's access. In Claude, go to Settings →
Connectors → Add custom connector and paste https://adlibraryspy.com/api/mcp.
Any other agent: paste "Read https://adlibraryspy.com/SKILL.md and follow it to research my competitors' Shopify stores and Meta ads."
All clients and tools →

Teams, API and extension

• Free team workspaces with roles, invites, shared saves and an activity log. Sign-in is passwordless (an emailed 6-digit code or magic link, or Sign in with Google)
• API keys for the MCP endpoint (Settings → API)
• Installable as an app on your phone or desktop
• In-app feedback (sidebar → Feedback): stored in the feedback table and emailed to FEEDBACK_EMAIL, with Reply-To set to the sender (10 messages an hour per user; FAIR_USE.feedback). Rate limits for agents: /SKILL.md
• A Chrome extension that shows any Shopify store's traffic, ads, products and apps in one click

Why trust the numbers

• A figure appears only if something measured it, and it carries its source and period, for example
  SimilarWeb · Aug 2026. Traffic history shows only real months, and a chart needs two real data points
  to be drawn at all.
• A missing value is shown as nothing, never as 0, a dash or a guess.
• AI labels are marked as model judgments, with a confidence. When a label is missing, the model wasn't sure.

The full rules are in docs/data-honesty.md.

This week's Shopify breakouts

Updated every week by a GitHub Action from the public weekly report.

Week 39, 2026 · updated 2026-09-25 · full week · all weeks

Top scaling stores

_{Stores adding the most ads in the Meta Ad Library, with measured traffic behind them. Source: Change in running ads in the Meta Ad Library, AdLibrarySpy index (Sep 2026 snapshot). Traffic: SimilarWeb, Aug 2026.}

Fastest traffic growth

_{Largest month-over-month jump in measured visits, among stores that already had 20K+ visits. Source: SimilarWeb measured visits, Jul 2026 → Aug 2026.}

Run it yourself

git clone https://github.com/PSA-Team-source/AdLibrarySpy && cd AdLibrarySpy
npm install                    # Node 22.18+
cp .env.example .env.local     # set DATABASE_URL (PGSSL=off for a local Postgres) and SMTP_*
npm run migrate
npm run dev                    # http://localhost:4311
npm test && npm run typecheck

SMTP is required because every sign-in is a link sent by email. A local mail catcher such as Mailpit works
(SMTP_HOST=localhost, SMTP_PORT=1025).

Where the data comes from: accounts, workspaces, the brandtracker, saves, API keys, OAuth and the public
pages all run on your own Postgres. Store and ad data come from the AdLibrarySpy market index, a hosted
service this app is a client of. Its shop endpoints are public, so a local instance shows stores out of the
box. Ad creatives need a service account (PLATFORM_JWT_SECRET, MARKET_SERVICE_ACCOUNT_ID). Without one,
the ads screens stay empty rather than inventing anything. To run your own instance against the full index,
open an issue.

Configuration

All variables are listed, with placeholders, in .env.example.

Tech stack

Next.js 15 (App Router, server actions) · React 19 · TypeScript · Tailwind · Radix UI and shadcn/ui ·
TanStack Query · PostgreSQL · ClickHouse (optional funnel analytics) · Cloudflare edge cache.
Multi-tenancy is enforced in lib/auth/guard.ts: a workspace_id always comes from the session, never
from the request. See docs/architecture.md for the layout, the traffic model,
the AI labels and the MCP server.

Also in this repo

Contributing

Issues and pull requests are welcome. See CONTRIBUTING.md and our Code of Conduct. Questions and ideas go to Discussions. Adding app and pixel
signatures to shopify-inspect is an easy first contribution. If AdLibrarySpy helps you, a ⭐ helps other
people find it.

Star history

License

MIT. The AdLibrarySpy name and logo, third-party logos and the screenshots are not covered.
See NOTICE.md.