Privacy

The short version

We do not store the domains you look up, there is no account to create or log in to, and the only personal data we ever collect is an email address if you choose to subscribe to the newsletter — which goes to Kit and not to us — and anything an AI agent puts into a feedback report sent through our MCP server, where we ask it to include none. We use two analytics services, Google Analytics and DataFast, to see which pages get read — but only after you say yes to the banner shown on first visit. Neither script is on the page until you accept, and you can change your answer any time from "Cookie preferences" in the footer.

The one thing worth knowing before you use the tools: when you run a lookup, the domain you entered is sent to public DNS resolvers operated by other companies, because that is how a DNS query is answered. Details below.

What the DNS tools send elsewhere

A lookup on this site is performed by our server, which forwards your query to public DNS-over-HTTPS resolvers. The DNS Lookup, MX, SPF and DMARC tools query Cloudflare, falling back to Google if Cloudflare does not answer. The propagation checker deliberately queries four independent resolvers at once — Cloudflare, Google, AdGuard and DNS.SB — because comparing their answers is the entire point of that tool.

Those resolvers receive the domain being looked up and the network address of our server, not yours. Each operates under its own privacy policy, which we do not control.

We do not write the query to a database, a log file or any storage. Once the response is returned to your browser, we no longer hold it.

What we process but do not keep

To stop one client from overwhelming the resolvers, each lookup is counted against your IP address using Cloudflare's rate limiting. The address is used to increment a counter and is not stored by us or written anywhere we can read back.

We use Cloudflare as our infrastructure and security provider: it serves the site, blocks abusive traffic and applies the rate limiting above. In doing so it records standard request logs — IP address, user agent, requested URL — as any host does. We do not read those logs in the ordinary course of running the site and do not copy them anywhere; Cloudflare retains and processes them under its own privacy policy (https://www.cloudflare.com/privacypolicy/), and the fact that a provider does the processing does not remove our responsibility for choosing that provider.

Requests from known AI crawlers (not people) are reported server-side to DataFast so we can see which bots read the site. This is a count of crawler visits by bot name and path; it involves no visitor of ours.

The MCP server at inetgeek-mcp.palashbagchi.workers.dev is a separate service from this site, and an AI assistant rather than a browser is what talks to it. It counts three things: completed handshakes, with the product name the connecting client declares for itself; calls per tool, by tool name; and the User-Agent header, which names the client software and its version. It records no IP address, no other header and no argument — the domain you check with a DNS tool through it, or the provider you look up, is not written down.

The User-Agent is the one header kept, and only because a client identifies itself just once, when it connects: the server holds no session and issues no key, so every later call would otherwise be anonymous and the counts would describe nothing. It names software, not a person. Because of that same statelessness, the handshake figure counts connections rather than people — a client reconnecting after a restart is indistinguishable from a new one. Cloudflare logs requests to this server as it does for the site.

Cookies and tracking

This site uses two analytics services to see aggregate traffic — which pages get read, roughly how visitors found them, nothing else: Google Analytics (GA4) and DataFast (datafa.st). Both are off by default. The banner shown on your first visit asks first, and neither script is loaded until you accept: decline, or take no action, and nothing loads and no cookie is set. Accept, and both load on that page and on every later visit from the same browser. Change your mind later from "Cookie preferences" in the footer of any page; declining there stops both from collecting anything new, and a script already on the current page is fully gone on the next page you open.

Once accepted, Google Analytics sets cookies (typically _ga and _ga_*) that hold a randomly generated ID for your browser, used to distinguish one visitor from another across a session and across return visits. We do not set any cookies ourselves outside of what Google Analytics sets when you accept.

Google Analytics does not log or retain your IP address as a stored field — Google discards it after using it to derive an approximate location (city-level at most). Google processes the data it collects under its own privacy policy, not ours, since Google is the one storing and processing it; see https://policies.google.com/privacy for what that covers. You can also opt out of Google Analytics tracking across all sites using Google's browser add-on: https://tools.google.com/dlpage/gaoptout.

DataFast is a web-analytics service that reports page views and referrers to us in aggregate. It runs only after you accept, alongside Google Analytics, and processes what it collects under its own privacy policy: https://datafa.st/privacy-policy.

A small traffic widget from the same service, DataFast, appears in the page margin on wide screens — an embedded chart, not a script we control. It loads on the same acceptance as the analytics above: nothing is requested from datafa.st for this widget until you accept, and declining or withdrawing consent means it is gone on the next page you open, the same as the two scripts.

Comments are hosted by EchoThread (echothread.io) and appear under provider, comparison, category and stack pages. Like the analytics above, the widget is off by default and nothing is requested from echothread.io until you accept — the script is injected on acceptance, not shipped in the page. Once it loads it fetches the thread for that page and can send usage beacons to api.echothread.io. If you sign in to comment, you do so through Google, GitHub, X or Facebook, and EchoThread holds that account identity and your comments under its own privacy policy, not ours. Reading this site never requires signing in, and declining leaves the comment area as a short notice with a button.

We do not use advertising cookies, a tag manager beyond what ships GA4 itself, or any session-recording or heatmap script.

The one third-party request unrelated to analytics is to logo.dev, which serves the provider logos shown on provider pages. We use it because the marks are recognisable where initials are not; the request happens automatically when a page with logos renders, and loading an image from any third party discloses your IP address and the referring page to them. Logo.dev processes that under its own privacy policy: https://logo.dev/legal/privacy. If we stop using the service, provider initials are shown instead and no external request is made.

If you subscribe to the newsletter

There is a newsletter sign-up at the foot of every page. It is optional, and nothing on this site requires it.

Subscribing sends one thing to one place: the email address you type goes to Kit (kit.com), the service that stores the list and sends the mail. It does not touch a database here, because there is not one — our server passes it straight to Kit and keeps no copy, no log line and no cookie. Kit is a US company and processes the address under its own privacy policy.

Sign-up is double opt-in. Kit sends a confirmation email and nothing else is ever sent unless you click it, which means someone typing your address into the form cannot subscribe you. Every issue carries an unsubscribe link, and unsubscribing is immediate and permanent unless you sign up again.

The form asks for an email address and nothing else. We do not ask for a name, we do not enrich the address against anything, and the list is never shared, sold or used for anything but the newsletter it was collected for.

Shareable lookup links

Tool results can be shared as a link containing the domain, for example /tools/dns-lookup?domain=example.com. That link contains whatever you typed, so treat it as you would any URL before sharing it.

These URLs are excluded from search engines in our robots.txt, and the canonical version of each tool page is the plain page without a query.

Feedback sent through the MCP server

Our MCP server has one tool that stores anything: send_feedback, which an AI agent can use to report a wrong figure, a broken tool or something missing. We keep exactly what the agent submits — the kind of report, the message, and optional fields naming a tool, a provider, a criterion and a supporting link — together with the time and the User-Agent of the client software, which names a program and not a person.

We do not keep the IP address. It is used to count how many reports one client sends in a minute, so the tool cannot be flooded, and is not written anywhere. The tool asks agents not to include personal data, credentials or their user's code; if one does anyway, it is stored as sent, which is why reports are deleted on a schedule: 90 days after we close a report, and one year after it arrives whether closed or not. A daily job enforces both.

Reports are read by the person who maintains the site and are not published. A report that leads to a corrected figure shows up in the ledger as a correction to that figure, without the report's text.

Your rights

We do not keep a user-account database, a lookup history, or any record that identifies a visitor in our own application systems, so there is nothing there to retrieve, correct or delete. Our providers may nevertheless process limited technical information to run and secure the site, and this section says who holds what and how to reach it.

Our systems: nothing stored, apart from MCP feedback reports; tell us the report id the tool returned, or what it said, and we will delete it. Cloudflare: request logs, held by Cloudflare under its own policy; send a request to us and we will assist with what we can see and forward what we cannot. Kit: your email address, if you subscribed; unsubscribing from any issue removes it, and Kit's privacy policy governs anything further. Google and DataFast: analytics data collected only after you accepted; declining or withdrawing consent from the footer stops both collecting anything new, and each provider's own privacy policy governs what it already holds from before you declined, if anything.

For visitors in the EEA and UK: the legal basis for analytics is your consent, given on the banner and withdrawable from the footer; for the newsletter it is your consent, given by double opt-in and withdrawable by unsubscribing; for request handling, rate limiting and MCP feedback reports it is our legitimate interest in operating, protecting and correcting the site. Google, DataFast, Kit and Cloudflare are US companies and may process data in the United States under their own transfer safeguards.

Who is responsible

inetGeek is operated by Immortal Reality PA LLC, 6375 Penn Ave Ste B, Pittsburgh, Pennsylvania 15206, United States. Contact us through immortalai.us.

If this policy changes, the change will be made in the same update as the change to the site that caused it.