Privacy policy

We resolve wallet addresses to the social accounts attached to them. To do that we hold your email address, what you have paid, what you have looked up, and enough technical data to keep the service standing up.

We do not sell your data, we do not run advertising, and we do not show your lookups to other customers. The addresses you look up can go to the companies that resolve them for us, described below, and a strip on our home page shows the size and match rate of some large lookups, with nothing that identifies them. We do keep the wallet-to-identity mappings a lookup discovers, and the section on addresses you look up explains exactly what that means, because it is the one thing here a reader would not guess.

Our terms of service say what customers may do with what we return, and what they may not do to the people in the index.

Your account. An email address, which is the whole account: there is no password and no profile. It exists so a sign-in link can reach you and so credits can belong to somebody. An account created by paying onchain has a wallet address instead, and no working inbox behind it. A new account also gets up to five welcome emails over its first two weeks, the last of which asks you to buy a pack, and an account that has not bought may get one plain check-in email, sent automatically in a person’s name. Every one carries an unsubscribe link. We keep a record of which ones we sent, and a copy of each check-in in our support inbox.

What you paid. For a card payment, the customer and payment references our payment processor returns, and the date. Card numbers never reach us: the payment happens on the processor’s own pages. For an onchain payment, the paying address and the settlement reference, both of which are already public on the chain.

Sanctions checks. Before we accept an onchain payment, we check the paying address against the US Treasury’s list of Specially Designated Nationals, and we keep a record of the check: the address, the date of the list, the result and when it was made. If an address that paid us is added to that list later, we suspend the account the payment was for and note on the account why. At checkout we also read the country and region your connection comes from, as our host reports it, so that we can refuse a sale where sanctions law forbids one. We do not store it.

What you looked up. The addresses you submitted, any other columns in your file, the name you gave the list and, if you imported it from a contract, which contract, the results, and a count. Saved lookups keep the full result set, your columns included, so you can open them again. Saving is on by default, and you can turn it off before a list runs. When you look up the wallets behind an account, we save the handle you searched for with the results, under a name that includes it. If you are not signed in, a lookup is tied to an identifier your browser keeps rather than to an account.

How the service is used. Page views and product events (an upload started, a checkout reached, a limit hit), each carrying a browser identifier and sometimes your email address, and a yes or no for whether it came from a known crawler or an automated browser. That is worked out from the browser string, which is not kept, and from whether your browser reports that software is controlling it. When the answer is yes, the event also records which crawler it was, or that the browser was automated. Also a record of each request an API key makes: the endpoint, how many addresses it carried, the response status, the credits and the time, never the addresses themselves, so rate limits and credits can be counted.

Where you arrived from. Recorded once per browser, the first time you land: the domain of the site that linked you, and any campaign tag on the address you followed. Never the full web address you came from, because those routinely carry other people’s search terms and session tokens, which we do not want and do not keep. If you create an account, that one short value is stored with it so we can tell which campaigns brought people. It is never updated afterwards.

If you use X from here. Building an X list from your results asks you to sign in to X. We hold a sealed token that can write lists on your account, and the accounts you chose to add, only until the list is built; then both are deleted. The job’s record (the list name and description, your X account name and id, the list’s id on X and the counts) stays. A list you never authorize is emptied at the next daily cleanup. Claiming an address also asks you to sign in to X, and the section on the index below says what a claim keeps.

Technical data. Your IP address, held only as a counter: against an hourly bucket, so an endpoint cannot be scraped, and against a daily one, so the free allowance for visitors who are not signed in can be counted. We do not store the browser string of the device you sign in on. Our host keeps request logs for no longer than 30 days. They record the web address of each request and the browser that sent it, and for a single-address or reverse lookup through the API that web address contains the address or the handle. The messages our own code writes to those logs mask email addresses and wallet addresses.

This is the part worth reading twice, because it is how the product works rather than an aside.

When you run a list, on this site or as a job through the API, every address we hold no fresh answer for is sent to the third-party identity data providers described under “Who else sees it”. A deep scan on an account with credits also reads onchain ENS records, which sends the address to a blockchain data provider. This happens whether or not you are signed in, and the other columns of your file are never sent to them. A fast scan answers from our index and our cache of recent results alone and sends nothing, whatever the size of the list, and so does a single-address or batch lookup through the API.

What comes back is written to a permanent index, and that index then answers other people’s lookups of the same address. Each record also keeps a counter that goes up whenever a lookup, or one of our own refresh jobs, writes to it. It never records who ran the lookup. When a list finds that an address resolves to nobody, we record that too, and a later list does not re-ask about it for 30 days.

What is shared this way is what the sources said about the address: which X handle, Farcaster account, ENS name, Lens profile or GitHub account it belongs to, a second X account the owner also published, the follower counts of those accounts, whether the X account still reaches anyone, whether the wallet appears to belong to an AI agent and what kind, the kind of evidence behind the link, and when we last checked. For an address that resolves to nobody, other customers can see when we last checked it. Those facts came from public sources, and they are what we sell.

What is never shared is anything about you: that you ran the lookup, which addresses you submitted together, what your list was for, or what any of it told you. Your lists are yours, and no customer can see another customer’s. The one exception is a strip on our home page of recent large lookups: how many addresses a list had, how many matched, and when it finished, with no name, account or address attached. Any completed list of 25 or more addresses from the last seven days with a match rate above 8% can appear there, our own imports of collections included. If you would rather yours did not, write to help@walletlink.social and we will hide it by hand.

Raw results are also cached for 7 days, so a repeated lookup of an address we found something for does not ask the outside sources again in that window. It is still billed like any other lookup.

You may be in it without ever having used the service. Every mapping we hold came from a public source, and most links were published by the wallet’s owner: a Farcaster verification, an ENS or similar name record, a profile set by signing with the wallet, or a record we checked by hand. A name record or profile can name any X handle, and nobody checks that the handle’s owner agreed, so your handle can appear beside a wallet you have never used. Some links come from a third-party index that pairs a wallet with an account when nothing shows the owner made the link. We label those as aggregated wherever we return them, never as attested by the owner, and a search for the wallets behind an account never returns them.

We collect nothing from a private source. The index itself holds no wallet balances, no amounts and no transaction history, and no email address, phone number or postal address for anyone in it. Beside the mapping it does hold: which token and NFT collections some wallets have held; whether a wallet carries a public onchain attestation that an exchange verified its owner; an X account’s display name, follower count and whether it still reaches anyone; a Farcaster follower count; whether a wallet appears to belong to an AI agent, from a curated list or from words in its Farcaster bio; and the history of handle changes we have seen. A customer’s own file can carry other columns about the addresses in it, such as a balance, and a customer’s saved results keep the Farcaster bio we found; we keep those only in that customer’s lookups, as described above. A handful of well-known people appear on our home page as examples, by name and with their profile photos, X and Farcaster accounts and the wallets we have indexed for them.

Write to help@walletlink.social with the address or handle and we will remove it. We do not require you to prove ownership first, because the alternative is asking a stranger for more information than we already hold about them. Name every identifier you want gone. We act only on the ones you name, and we do not search the index for others that might be yours. The suppression list stores each identifier on its own row, with nothing that ties the ones in your request together. We keep the email thread, your message and our replies, for 90 days after the removal is done, so that we can apply the removal again if we ever restore our database from a backup, and then delete it.

What happens next. A person reads your email and runs the removal by hand; there is no form and no automation between you and it. Each identifier you name is deleted from the index and added to a suppression list that every write path checks, so an automated sweep that finds the same public record later cannot put it back. A removal lasts until you ask us to undo it, with the same proof that the address or account is yours that we ask for before sending a copy of what we hold (see “Your rights”). In the first 30 days we can also undo a removal made in error, as described below. Where a customer’s saved lookup that we still hold carries the link, the link is removed from it: everything we resolved for the identifier is stripped, so the entry reads as if nothing was found. An address that was part of the customer’s own uploaded list stays in that list, carrying nothing. A saved lookup whose subject is the identifier itself (a search for the wallets behind your handle) is deleted whole. We complete this within 30 days, the same period as every other request on this page.

A few copies survive, on purpose or for a limited time. For 30 days after a removal we keep what was deleted in a table read only by the removal tooling itself, kept so a removal made in error (a mistyped address, somebody else’s handle) can be undone. After 30 days that copy is deleted automatically. Our encrypted nightly backups include saved lookups and our list of AI agent wallets, and each is kept for 90 days, so a removed link can survive in a backup until then. If we ever restore one, we keep today’s suppression list rather than the backup’s, so nothing removed since that night comes back. An API batch answer kept for a retry is amended or deleted in a removal, and a replayed answer is filtered, so a retry never brings the link back. If the address was ever claimed on our claim page, the removal clears the claim record too, as a withdrawal on the claim page does. And the identifier you named stays on the suppression list, and on internal do-not-recheck lists that hold the identifier on its own, never the link.

If you can sign for the address, you do not have to wait for us. Claim your address works without a person in the middle, for an X account. You sign in with your email, sign a message with the wallet and sign in to X. Where we hold nothing for the address, your claim fills it, and where we hold the same account, it confirms it. Where we hold a different account, we record that you disagree and keep serving ours until the handle we hold stops reaching anyone, because a signature proves control of a key, and keys are lost and sold. A claim is tied to your walletlink account and keeps the address, your X account name and its numeric id, your signature, and a record of which version of the consent text you agreed to.

Once you have claimed an address, you can withdraw it from the same page. Withdrawing deletes what the index holds for the address and adds it to the same suppression list, so a later sweep cannot put it back. We keep a record on your account that you claimed the address and withdrew it: the address and a one-way hash of the X account id, so the same account cannot claim the free matches twice. A withdrawal also sends our support inbox a short email naming the address, a reference to your claim and the time, so that we can apply the withdrawal again if we ever have to restore our database from a backup; we delete that email after 90 days, when no backup older than it is left. Email stays the route that asks nothing of you, and it is the only route for a handle, for an address whose key you no longer have, and for asking to be removed from the index entirely.

Two things are beyond our reach whatever we build. A customer who exported a result before you asked still has their copy, and we cannot reach it. And a search engine may hold a cached copy of a page for a while after we change it.

The copies we hold ourselves we amend, except the ones listed above. Every list a customer runs keeps its addresses for 30 days in its job record, and a saved lookup keeps them until its owner deletes it, so for those copies we do know which account held the link. We do not tell those account holders that a removal changed their results, for three reasons. Telling them would reveal that the person behind the link asked to be removed. Tracing everyone who ever received a link takes effort out of proportion to what it would achieve, and a list run while signed out is tied only to a browser, so there is nobody to tell. And the removal reaches them anyway: every job result and saved lookup we hold is checked against the suppression list each time it is opened, so the link is gone the next time its owner looks. A single-address or batch lookup through the API leaves no address in our database: our record of API calls keeps the endpoint and the number of addresses.

We use other companies to run the service. Each receives only what its job needs, and each company’s own terms and privacy policy govern what it does with it.

• Vercel hosts the site and provides its page-view analytics.
• Neon hosts the database.
• Inngest ran lists of more than ten addresses in the background until 25 September 2026, and no longer processes anything for us. It kept a history of each of those runs that includes what the job held: the addresses, the other columns of your file, the results, and the account or browser that ran it. Neither our 30-day limit nor a removal reaches that history, which Inngest keeps under its own terms.
• GitHub runs our scheduled jobs and stores our nightly database backups. The backups are encrypted before they leave the job, with a key GitHub never holds, and each is kept for 90 days.
• Stripe takes card payments, holds the card details we never see, and receives your email address so the payment belongs to your account.
• Resend sends sign-in links, account mail, and the welcome and check-in emails.
• Cloudflare serves the domain and runs the assistant on this site and on the documentation site, which sees the questions typed into it.
• Google Workspace hosts our support mailbox, so it holds the email you send us and our replies, including removal requests for as long as we keep them.
• Mintlify hosts the documentation site.
• X receives the accounts you add to a list you build here, through your own X account, and tells us your account name and id when you build a list or claim an address.
• Warpcast receives the direct messages you send to Farcaster accounts from your results, sent with your own Warpcast API key. The key and the message pass through our server on the way and are not stored there. If you choose to save it, the key stays in your browser until you remove it.
• PayAI, a payment facilitator, settles onchain payments. It sees the paying address and the amount, both already public on the chain.
• Third-party identity data providers receive the addresses in the lists you run that we have not answered recently, the Farcaster usernames in your results when we look up their account numbers, and the X handles whose status we check. A blockchain data provider receives addresses when a deep scan reads onchain ENS records. They are named by category rather than individually, which is what a controller is permitted to do and what keeps our sourcing from being a public price list for anyone copying the product.

We will also hand over data where the law requires it, and we would tell you unless we were forbidden to. If the business is ever sold, what we hold moves with it and this policy travels with it too.

For your account, your payments and the lookups you run, we use your data because you asked for the service and it cannot run without it. We keep payment records because tax law requires them, and we check onchain payments against the sanctions list and keep a record of each check because sanctions law requires us not to deal with the people on it. We keep IP counters and product analytics because we have a legitimate interest in keeping the service standing. We send the welcome and check-in emails to account holders on the same basis, a legitimate interest in telling a new account what it can do, and every one of them carries a one-click unsubscribe.

For people in the index, who never signed up, we rely on legitimate interests. We help projects reach people who publicly linked a wallet to a social account themselves, and we show how each link was established. You can object at any time. Write to help@walletlink.social and we will remove what you name, as described above.

A cleanup job runs once a day and deletes what has passed its period, so an item usually outlasts its period by up to a day, and a large backlog can take a few days more. Items kept until you act, or kept indefinitely, say so rather than implying an expiry.

One cookie, wts_session, set when you sign in and readable only by the server. It is what keeps you signed in, it lasts 30 days, and signing out deletes it. There is no advertising cookie and nothing here tracks you across other sites.

Your browser also keeps a few things for this site in its storage: an identifier that tells one visit from the next and ties lookups you run while signed out to this browser, a per-tab identifier used to group events from a single visit, where you first arrived from, your email address while you are signed in, the lookup in progress, your theme, a few interface settings and, if you chose to save it, your Warpcast API key. Clearing your browser storage clears all of them. You lose access to lookups you ran while signed out, and nothing else breaks.

Wherever you are, you can ask us for a copy of what we hold about you, ask us to correct it, ask us to delete it, ask us to stop using it, or ask for it in a portable form. Depending on where you live, some of those are rights rather than requests; we do not distinguish, because answering everyone the same way is simpler than deciding who is entitled to what.

If you are in the index rather than a customer, we will ask you to show that the address or account is yours before we send a copy of what we hold for it, because the same answer sent to a stranger would hand them the link you may want hidden. We accept either of two proofs: a message signed with the wallet, or a public post from the social account that names your request. With that proof, you get a full answer. Removal never needs that proof, and our reply to a removal request never confirms whether we held anything.

Write to help@walletlink.social. We answer within 30 days, and we do not charge for it. If you are in the UK or the EU and we have not resolved something, you can complain to your data protection authority.

Some things we cannot delete on request. Payment records are kept for at least 7 years, for tax and accounting, and records of sanctions checks for 5 years, to show that we complied with sanctions law. Deleting your account does not remove a wallet-to-identity mapping from the index, because that mapping is not about you unless the wallet is yours, in which case the section above is the one that applies.

Everything travels over HTTPS. API keys and sign-in tokens are stored as SHA-256 hashes, never as the value itself, which is why a key is shown exactly once and cannot be recovered afterwards. An application connected through OAuth holds an access token that expires every hour and can be cut off from your account at any moment. Our scheduled GitHub workflows and the nightly backup reach the database through limited roles. The site itself, including the jobs it runs on a timer, connects with full access, and the removal quarantine is closed to every other role.

No system is perfect. If something goes wrong that affects you, we will tell you, and we will tell the relevant authority where we are required to.

This is a product for business and professional use, and it is not for anyone under 18. We do not knowingly hold data about a child. If you believe we do, write to help@walletlink.social and it will be removed.

When this changes, the date at the top changes with it, and the change is recorded in the public changelog like everything else. For anything that materially affects what we do with data we already hold, we will email account holders before it takes effect.

help@walletlink.social, for a question, a request, or a complaint. It reaches a person rather than a queue. walletlink.social is operated by Starl3xx Labs LLC, a limited liability company organized in Wyoming.