Privacy Policy
Effective date: 25 October 2026
Sitewell: Link & Sitemap Check is published by Frostpine labs ("we", "us"). This policy explains what Sitewell: Link & Sitemap Check does with data when you install it on a Shopify store. It applies to merchants: the store owners who install the app and the staff they give access to. It also applies to website owners who contact us about the link checker.
The short version
- Sitewell: Link & Sitemap Check only reads your store. It cannot change anything in it.
- It does not request access to your customers' data or your orders.
- It stores your shop's domain, the access tokens Shopify gives the app (encrypted), and the results of its scans.
- To check links to other websites, it sends a request to each website your content links to, except websites that asked us not to. It reads only the status of the answer, not the page.
- It requests a few public files from your own storefront (
robots.txt, your sitemaps,agents.mdandllms.txt) and saves only the result of each check, not the text. - It has no advertising or analytics service, and it sets no tracking cookies.
- When you uninstall the app, we delete your data from our database at once.
What the app can see in your store
When you install the app, Shopify asks you to approve three read-only permissions:
read_products: products and collections. The app reads their titles, handles (the last part of their web address), descriptions, whether products are published, and when each was last updated.read_online_store_pages: pages, blogs and articles. The app reads the titles, handles, body text, publish status and last-updated time of pages and articles, and the handles of blogs.read_online_store_navigation: navigation menus and URL redirects. The app reads menu names, menu item titles and links, and the "from" path of each redirect.
The app also reads basic facts about the shop itself: its primary domain, its myshopify.com domain, and whether it is a development store. To work out your plan, it asks Shopify which subscriptions of this app are active (their name and status). When you open the Sitemap & AI page, it also reads your store's name once, to fill in the agents.md template shown there. It does not save the name.
The app does not request access to customers, orders, payments, staff accounts, store policies or theme files. It never writes to your store, and it adds no code to your theme.
What we store, and why
- Your shop and its sign-in. Your shop's myshopify.com domain, plus the access token and refresh token the app receives from Shopify, their expiry times, the permissions you approved, and a technical sign-in value. Both tokens are encrypted before they are saved (AES-GCM, with a key kept as a secret in our hosting account, not in the database). We use this so the app can read your store with your permission, including for scheduled scans when you are not signed in. The app asks Shopify for an offline sign-in only. The app does not hold staff user IDs. We never keep a name or email address of a staff member.
- Your plan. Whether your store is on the Free or Pro plan as Shopify reports it, when we last checked (we reuse the answer for up to 10 minutes), when the app was installed, the date of your last automatic scan, and the time of the last failed attempt to start one. We use this to apply your plan's limits and to schedule daily scans for Pro stores. Payments are handled by Shopify, and we never see your card details.
- Scan history. For each scan: when it started and finished, whether you or the daily schedule started it, the plan it ran under, its status, its score from 0 to 100, how many resources it found and scanned, a short message if it failed, and a technical count of the database rows it wrote. We use this to show your score and its trend.
- Scan results. For each problem found: the kind of problem, the title of the product, collection, page, article or menu where the link sits, a link to that item in your Shopify admin, the link's web address as written in your content (without any part after a "#", and for links to other websites with tracking parameters and any username or password removed), and a short description. For a menu, the description lists the titles of the menu items that hold the link. For a link to another website, it also gives the reason the check failed (for example "HTTP 403" or "timeout") and the titles of up to three items where the link was found. We use this to show you exactly where each problem is. For the sitemap,
robots.txtand AI-file checks, the results are the web address of the file and a short description of what was found. - Working data during a scan. While a scan runs, the app keeps a temporary map of your store: the handles of your products, collections, pages, blogs and articles, whether each is published, when each was last updated, your redirect paths and your menus (names, item titles and links). It also keeps the list of items to scan and the list of other websites' web addresses to check, with the number of places each is linked from, up to three sample places (the item's ID, type, title and admin link), and the result of each check (status code, final address, any error, and the time of the check). We use this to run a scan in small steps.
- Scan progress records. Each scan also leaves small progress records, one per step of the scan, kept with the scan's history record. They include your store's domain names, the titles, handles and product counts of up to 10 of your largest collections, the handle and ID of your contact page if you have one named
contactorcontact-us, and, for the storefront check, the web address of each file checked and a summary of what was found (the status, whetherrobots.txthas aSitemap:line, which well-known AI crawlers it blocks, whetheragents.mdlooks like Shopify's default). They hold no file text. - A log of websites contacted. The domain name of each website the app has contacted, and when it last did so. We use it only to keep to the rate limit described below. It is shared across all stores and has no shop column, but when the app requests the files on your own storefront, your storefront's domain name appears in it like any other.
- A list of websites that asked not to be checked. The domain of each website owner who asked us to stop (see "Opting out" below), the date we added it, and an optional internal note. It is not tied to any shop.
- Technical logs. Our hosting provider, Cloudflare, records standard logs of requests to the app. These can include technical details such as the IP address and the web address requested, which can contain your shop domain. Our own log lines record events such as a webhook arriving for your shop domain, a failed sign-in refresh for your shop domain, or a job failing, and they can carry the text of an error. They do not include the content of your store or any token. We use logs to keep the service running and to fix faults. Cloudflare keeps these logs for 3 days on the plan the app uses.
- Support emails. If you email shopmanager456@pm.me, we receive your email address and your message. We use them to reply to you.
What we do not collect
- Customer data. The app does not request access to customer or order data, so it cannot read your customers' names, emails, addresses, orders or payment details.
- The text of your store. The app reads descriptions and page bodies in memory to find links, then discards them. It does not save them. It saves only the titles and web addresses described above. If a title or a link contains personal information, that information will appear in your scan results until they are deleted.
- The text of your storefront files. The app reads
robots.txt, your sitemaps andagents.mdin memory, up to a size limit for each, to check them, and saves only the result. It does not read the content ofllms.txtat all; it only looks at whether the file loads. - Staff details. The app stores no names or email addresses of your staff.
- Email addresses and phone numbers in links. Links that start with
mailto:ortel:are ignored. - Advertising or analytics data. The app uses no advertising or analytics service and sends no data to one.
How we use your data
We use your data only to run the app for you: to scan your store, show the results, apply your plan's limits, bill you through Shopify, keep the service running and secure, and answer your questions. We do not sell your data, and we do not use it for advertising.
Where your data is stored and processed
- Sitewell: Link & Sitemap Check runs on Cloudflare Workers. Its database is Cloudflare D1, which is Cloudflare's database service.
- Cloudflare runs Workers in data centres around the world, so your data may be processed outside the country where you live. The app's database is hosted by Cloudflare in eastern North America (the United States or Canada).
- Shopify supplies your store's data through its Admin API, and the app is shown inside your Shopify admin. When you open the app, your browser loads Shopify's own scripts and a font from cdn.shopify.com. The public Privacy, Support and bot pages load no script, font or other outside file.
- We do not copy your data to other servers or services.
Who else the app contacts
- Shopify. The app reads your store through Shopify's Admin API, read-only. It also uses Shopify's billing system for paid plans, so Shopify sends the app notices when your subscription changes.
- Cloudflare. Our hosting and database provider, as described above. For a link that gets Cloudflare's own 530 error page, the app also asks Cloudflare's public DNS service (cloudflare-dns.com) whether the website's name exists, sending only that name.
- Websites your content links to. The app sends a request to the other websites your content links to, to see whether each link works. The next section explains exactly how.
- Your own storefront. The app requests a few public files from your store, using the same User-Agent as the link checker:
robots.txt,sitemap.xml,agents.mdandllms.txt, and up to 10 of the sitemap files that yoursitemap.xmllists. These requests go only to the store's own addresses (its primary domain, the www or non-www version of it, and its myshopify.com address); a redirect anywhere else is noted and not followed, and a listed sitemap on another website is noted but not requested. It follows up to 5 redirects per file. It reads at most 32 KB ofrobots.txt, 32 KB ofsitemap.xml, 16 KB ofagents.mdand 8 KB of each listed sitemap, and none ofllms.txt. It saves only the result of each check. If your storefront redirects to its password page, or answersrobots.txtwith a 401 or 403, it treats the store as protected and skips the rest of these checks.
No one else. The app does not use any analytics, advertising, email, error-tracking or AI service.
How the link checker treats other websites
- What it requests. One request for each different web address found in your content, including the addresses of images and other embedded media. It first sends a HEAD request, which asks only for the status. If the website answers a HEAD request with 403, 404, 405, 410 or 501, it sends one normal GET request to the same address to confirm, and cancels the download without reading it.
- What it reads. Only the status code, a few response headers, and for a redirect the new address. It never reads a body, and it does not keep or follow anything on the page. It follows at most 5 redirects and waits up to 8 seconds for an answer.
- One name lookup for a missing website. If a link gets Cloudflare's own 530 error page (often a website name that does not exist), the app makes one lookup at Cloudflare's public DNS service (cloudflare-dns.com) to check whether the website's name exists. Only the name is sent, with the app's User-Agent, and only the short answer is read.
- How often. It leaves at least one second between requests to the same website, and sends at most 10 requests to one website in each short step of the scan. In rare cases, when the scans of two stores overlap, a website could get two requests within one second.
- Who it says it is. Each request carries the User-Agent "SitewellBot/1.0" followed by a link to the app's public bot page, so website owners can see who is asking.
- What the website sees. The address requested, our server's IP address, the User-Agent and a generic Accept header. We send no cookies, and we do not tell the website which store the link came from.
- What it avoids. Before it stores or requests an address, it removes common tracking parameters (
utm_*,fbclidandgclid) and any username or password. It refuses addresses that are not http or https, addresses that are not public IP addresses, names such aslocalhostand names ending in.localor.internal, and ports below 1024 other than 80 and 443. A public-looking name that someone has pointed at a private address is not detected by this check. - Your own store. Links to your own store are not requested over the web. The app checks them against the list of products, pages and other items it read from Shopify.
- Opting out. Website owners can ask us to stop checking links to their site by emailing shopmanager456@pm.me with the domain name. We add the domain to an exclusion list. From then on the app sends no request to that domain or any of its subdomains, including when a redirect leads there, and the merchant's scan shows the link as "couldn't verify" because the site owner asked not to be checked. We aim to add the domain the same day, and always do so within 7 days. The checker does not read other websites'
robots.txtfiles.
How long we keep data
- Sign-in tokens, shop record and plan details: until you uninstall the app. Tokens are replaced each time Shopify refreshes them.
- Scan history, including the progress records: your 30 most recent scans. Older scans are deleted in the background.
- Scan results: the results of your 3 most recent completed scans, plus any scan that is still running. Older scans keep only their history record, with no results.
- Working data: the temporary map, the lists and the check results are deleted in the background, in small batches, after a scan finishes.
- Log of websites contacted: entries older than one day are deleted in the background.
- List of websites that asked not to be checked: until the website owner asks us to remove the entry. We do not keep the requester's email address in readable form: the entry holds only a one-way hash of it, so that we can check that a later removal request comes from the same person.
- Cloudflare logs: 3 days, set by Cloudflare.
- Support emails: up to 1 year, then deleted from the support mailbox.
Deleting your data
When you uninstall Sitewell: Link & Sitemap Check, Shopify notifies the app, and the app deletes everything it holds for your shop in one step: your sign-in tokens, plan details, scan history, scan results, progress records and working data. If the deletion fails, the app answers with an error so that Shopify tries again.
Shopify also sends a "shop redact" request after an uninstall (Shopify says about 48 hours later). When it arrives, the app runs the same deletion again to remove anything that is left. If nothing is left, nothing happens.
If Shopify passes on a data request or a deletion request from one of your customers, the app confirms it with a success answer and records only that a request arrived for your shop. The app holds no customer data, so there is nothing to send or delete.
The log of websites contacted is not tied to any shop, so it is not removed at uninstall. It expires on its own after about a day. Cloudflare's logs, the exclusion list and any support emails are not removed at uninstall either. They follow the periods above.
Cloudflare keeps a point-in-time history of the database for 7 days on the plan the app uses. Deleted data may remain in that history until it expires, after which it is gone.
If you reinstall later, you start from scratch. Your old scores and results are not restored.
Security
This is what the app does today to protect your data:
- It has read-only access, limited to the three permissions listed above.
- It checks the signature on every webhook it receives from Shopify, rejects any request with an invalid signature, and, where the signed content names a shop, acts only when it names the same shop as the request. Subscription updates name no shop; for those the app only marks the plan to be re-checked with Shopify.
- The app's pages send a Shopify session token with their requests to our server. The server checks the token and rejects requests without a valid one.
- Access tokens and refresh tokens are encrypted in the database. The encryption key and our Shopify API secret are stored as secrets in our hosting account, not in source code. Other fields in the database, such as your shop domain and scan results, are not encrypted by the app.
- The database is a private Cloudflare database that is reached only through our application.
- The link checker refuses private and internal addresses by the patterns described above. It cannot detect a public-looking name that has been pointed at a private address.
- The app is served over HTTPS.
No system is perfectly secure. If you think you have found a security problem, email shopmanager456@pm.me.
Cookies and browser storage
The app's own code does not set cookies or use your browser's local storage. Inside your Shopify admin it signs you in with Shopify session tokens, so it works without third-party cookies. The public pages set no cookies. It has no advertising or analytics cookies.
Your choices and requests
You can uninstall the app at any time. That stops all reading of your store and starts the deletion described above. You can also ask us to delete your shop's data without uninstalling. We do that within 30 days: your scan history and results are removed and the app starts again as if newly installed.
Who is responsible for your data. For your shop's account data (its domain, tokens, plan, scan history and results) and any support email you send us, Frostpine labs decides how that data is used and is the controller. Shopify is a separate controller of your store's data under its own privacy policy. If a title or link in your store content contains personal information about someone else, we process it only on your behalf, as your service provider, to show you your scan results.
Your rights. Depending on where you live, you may have the right to ask for a copy of the personal data we hold about you, to have it corrected or deleted, to restrict or object to its use, to receive it in a portable form, and to withdraw consent you gave. We do not sell personal data and do not use it for targeted advertising. To exercise a right, email shopmanager456@pm.me, including your myshopify.com domain or the email address the request concerns, so we can confirm it is yours. We answer within 30 days, free of charge, and tell you if we need longer and why.
Complaints. If you are unhappy with how we handled your data, you can complain to the data protection authority where you live, for example an EU member state's supervisory authority or the UK Information Commissioner's Office. We would appreciate the chance to answer first.
International transfers. Our database is hosted by Cloudflare in eastern North America (the United States or Canada), and Cloudflare may process requests in other countries. Where your local law requires it, these transfers rely on Cloudflare's data processing addendum (https://www.cloudflare.com/cloudflare-customer-dpa/), which includes the EU standard contractual clauses and the UK addendum. If you are in the EEA, UK or Switzerland, your data may be held in a country whose data protection rules differ from yours.
Changes to this policy
If we change this policy, we will update the effective date at the top of this page. If a change is significant, we will also show a notice on the app's home page at least 14 days before the change takes effect.
Contact
Questions about this policy, and requests about your data: Frostpine labs, shopmanager456@pm.me.