DEV Community: Muhammad Inshaal Shahid

Agentic commerce for Shopify: how to sell when AI agents do the shopping

Muhammad Inshaal Shahid — Thu, 29 Oct 2026 11:10:16 +0000

This is the first holiday season where AI agents don't just recommend products, they buy them. In the space of six weeks, Shopify opened checkout to browser-based agents, launched a single connection that lets any AI agent search more than 1 billion products and pay with Shop Pay, and started sharing sale prices with AI shopping channels. If you sell on Shopify, your next customer might never see your homepage.

The numbers behind the shift

65% of shoppers say they'll use AI for at least one shopping task this season
1B+ products in Shopify Catalog that agents can search
5× more AI-referred US orders for Stanley 1913 vs the same window last year
+276% AI-referred sales growth at Cottonique, a small hypoallergenic apparel brand

What Shopify launched (September to October 2026)

28 September: WebMCP for checkout. Browser-based AI agents can now complete checkout on eligible Shopify stores, including Shop Pay.
9 October: compare-at prices for agents. AI shopping agents now see when a product is marked down, not just the current price. You can control this in admin.
13 October: Shop MCP. One connection (mcp.shop.com) lets any MCP-compatible agent search Shopify Catalog, build carts and complete Shop Pay checkout. Search needs no API key.
Agentic Storefronts. Eligible stores can surface products in ChatGPT, Google AI Mode and Gemini, Microsoft Copilot and Meta, and track AI sales in admin.

Two rules matter for trust: the agent must show the shopper the total and payment method and get confirmation before paying, and orders placed by agents land in your normal Shopify admin like any other order.

How an AI agent buys from your store

Shopper asks: "A gift for my niece who loves sharks, under $50."
Agent searches Catalog: reads your product data, price, stock and reviews.
Builds the cart: picks size and variant at your store.
Shopper confirms: total and payment shown, one tap to approve.
Order in your admin: paid with Shop Pay, fulfilled like any order.

Why small brands can win this

Traditional SEO rewarded whoever ranked for the exact keyword. Agents work differently: the shopper describes a problem, and the agent looks for the product that solves it. Shopify's own example: "Nobody with eczema types 'hypoallergenic bra'." An AI that understands "my bra makes my skin burn" can find the right niche brand even if it never ranked on Google.

Shopify's 2026 holiday report found 71% of regular AI users are more likely to choose an indie brand this year. The catch: an agent can only repeat the case your store has already made. If your "why this product" lives only in a lifestyle video or a vague tagline, the agent compares you on price alone.

Old playbook (built for search)	Agent-ready store
Stuff exact keywords into titles	Titles that say what it is and who it's for
Key details hidden in images	Materials, sizes and specs as real text
"Premium quality" with no proof	Specific claims: "100% organic cotton, no dyes"
Shipping and returns buried in a footer page	Clear shipping, returns and delivery areas
Sizes in a PDF chart	FAQ answering real customer questions

The agent-ready Shopify checklist

Fix your product data first. Clean titles, complete descriptions, correct product type, variants, GTIN/barcodes and metafields. Shopify Catalog structures what you give it; it can't invent what's missing.
Write for problems, not keywords. Add a short "Who it's for" and "What it solves" section to each product.
Make policies plain text. Shipping times, costs, delivery countries, returns and warranty, readable on the product page.
Check your agentic settings. Review Agentic Storefronts and the compare-at price toggle in Shopify admin.
Keep stock and prices accurate. Agents read real-time data; an out-of-stock surprise kills the sale and the trust.
Earn mentions off-site. Stanley 1913 says agentic optimisation means appearing in other authoritative sources too: reviews, press, comparison articles.
Make the store fast. Shoppers who click through from an AI chat still judge you in seconds.
Watch the AI sales report. Shopify labels AI-referred orders so you can see which channels and products agents favour.

The bottom line

Agents won't replace your store, but they are becoming a new front door. The brands that win this season will be, in Shopify's words, "found by a machine and chosen by a person." That means clear data for the machine and a real reason to buy for the person.

Want your Shopify store audited for agent-readiness before Black Friday? I'll review your product data, policies and speed, and fix what holds you back. Start with my free website audit or tell me about your store.

Originally published on my blog.

Sources: Shopify developer changelog, Shopify News: the first holiday season of the agentic era, Shopify changelog, TechCrunch, Shopify Help Center

Hackers now use AI agents to break into websites: how small businesses stay safe

Muhammad Inshaal Shahid — Thu, 29 Oct 2026 11:08:17 +0000

For years, small businesses told themselves the same thing: "Hackers go after big companies. Nobody's going to bother with my website." In October 2026 that excuse ran out. Researchers watched a single criminal point autonomous AI agents at online shops and let them break in with almost no human help. The agents tried more than 1,200 sites and reached websites belonging to more than 100 companies, including John Deere, SharkNinja and Cummins.

The campaign in numbers

1,200+ websites the AI agents tried to break into
~$5 estimated average AI cost per attack task
~23 hrs average time to a successful break-in
~10% overall success rate across the tools observed

What happened

Security researcher Snorre Fagerland of NorForge was scanning the internet for exposed copies of Cairn, an open-source AI tool built for legitimate, authorised penetration testing. From early October he found around two dozen Cairn control panels being used maliciously, left open on the internet. That let him watch the attacks happen in real time.

The busiest group, which NorForge calls AIA0001, ran 14 panels linked to more than 1,100 attack projects, targeting shops and payment systems in the US, Japan, Australia and Canada. Its goal appeared to be lasting access to e-commerce and payment infrastructure. NorForge confirmed at least one live web shell on a commercial website, and Reuters identified several victims from the attacker's own records.

The affected companies said the damage was limited: Deere fixed the problem and saw no sign sensitive data was accessed, SharkNinja said a standalone product-support site was hit and contained, and Cummins was investigating a vendor-run website. Two smaller businesses also confirmed they were compromised.

How the AI agents attacked

Short instructions: the operator wrote under 100 Chinese characters of tactics.
Pick targets: ranked by traffic; browser automation checked if shops took orders.
Hunt weak spots: SQL injection, unsafe uploads, auth bypass, old code.
Plant a web shell: a hidden upload form for lasting access.
Steal later: payment data, keys and customer records.

Why this changes things for small businesses

The attackers didn't use new magic. They used familiar weaknesses: SQL injection, unsafe file uploads, insecure deserialization, authentication bypasses and outdated framework code. What changed is the economics.

As Detectify's CEO put it, AI is changing the economics of cyberattacks more than the techniques. When one attempt costs around five dollars and an agent will keep trying for hours without getting bored, attackers can afford to go after sites that were never "worth the effort" before. That includes your shop, your clinic's booking site and your restaurant's ordering page.

Before	Now
A human tests sites one by one	Agents run hundreds of projects in parallel
Skips small sites: not worth the time	Every site is worth ~$5 to try
Gives up after a few failed tries	Keeps going for ~23 hours on average
Works office hours	Works 24/7, with little supervision

The leak inside the leak

There was a second lesson. The exposed panels showed which sites were attacked, what weaknesses were found and which passwords and API keys were collected. NorForge ran a secrets scanner over the data and found hundreds of secrets, 11% still valid. It also found legitimate security firms that had left real client tests on unsecured panels. If you hire someone to test your site, ask how they protect the results.

How to protect your website: a practical checklist

Update everything. CMS, plugins, themes, PHP, Node.js and frameworks. Old code was the way in for many of these break-ins. If you run Magento, apply Adobe's hotfix for the critical StyleSmuggler flaw.
Lock down file uploads. Allow only the file types you need, store uploads outside the web root, and never let uploaded files execute. A stray upload form was the foothold these agents aimed for.
Put a firewall in front. A web application firewall such as Cloudflare blocks common attack patterns and bad bots before they reach your server.
Rotate leaked keys fast. Treat any exposed password, API key or token as compromised. Change it, don't just delete the file.
Turn on two-factor login for every admin account, hosting and email.
Know what you own. List every domain, subdomain and old landing page. Forgotten "test" and "support" sites, like SharkNinja's standalone support site, are easy targets.
Watch for changes. Get an alert when a new admin user, plugin or unknown file appears.
Keep tested backups off the server, so you can roll back in minutes, not days.

Layers that stop an AI agent

Firewall: blocks bad bots and known attack patterns.
Patched code: no easy, known holes to find.
Safe uploads + 2FA: no foothold, no stolen logins.
Alerts + backups: catch it early, recover fast.

The bottom line

AI hasn't made hackers smarter. It has made them tireless and cheap. The good news: the same basics still work. Sites that are patched, firewalled and monitored are far harder for an agent to crack, and most agents will move on to an easier target.

Want to know how your site would hold up? Run my free website audit, or send me the link and I'll check updates, uploads, logins and firewall settings and fix what's weak.

Originally published on my blog.

Sources: The Hacker News, Dark Reading, NorForge, Reuters