Exposure and public records
What a stranger can already see about your business online
Before anyone attacks a business, they read about it — and almost everything they need is already published, deliberately, by systems that have to be public to work. This is what each of those records says about you, why it exists, and which parts you can actually change.
The short answer
Four public records describe your business to anyone who asks: DNS, which names your web host, your mail provider and often the software you use; mail records derived from it, which say whether a stranger can send email as you; certificate transparency logs, which permanently list every hostname you have obtained an HTTPS certificate for; and the domain registry, which publishes your registrar, your renewal date and whether the domain is locked.
None of this requires touching your systems, and none of it is a flaw — each record is public because it has to be for the internet to work. You cannot withdraw them. You can make what they say accurate, retire what should no longer be there, and fix the one that lets a criminal impersonate you.
Where does public information about a business come from?
Three separate systems, each public for a reason that has nothing to do with you.
- DNS
- The internet's address book. It has to be readable by anybody, because every mail server that delivers to you and every browser that opens your website starts by asking it a question. There is no such thing as a private public DNS record.
- Certificate transparency logs
- Public, append-only records of every certificate a publicly trusted authority issues. They exist so a certificate wrongly issued for your domain cannot be used quietly — browsers expect to see the proof of logging. The security benefit is real, and the side effect is that every name you have certified is searchable.
- The domain registry
- A domain name is a registration, and registrations have public particulars: the registrar of record, when it was created and when it expires, the status flags, the name servers. Personal contact details are usually redacted now; the operational facts are not.
What follows uses baharicourt.example — a fictional coastal hotel on the
reserved .example domain, and the same one our
worked sample report is written about.
What does DNS reveal?
-
baharicourt.example. IN A 41.203.0.10
Where the website lives — and therefore who hosts it, which is one lookup away from a support desk somebody can social-engineer.
-
baharicourt.example. IN MX 10 route1.mail-provider.example.
Who handles your mail. This single line tells a stranger which login page to imitate when phishing your staff.
-
baharicourt.example. IN NS ns1.serengeti-dns.example.
Who runs your DNS. A second supplier to impersonate — and the account that, if taken over, moves your website and your email at once.
-
baharicourt.example. IN TXT "google-site-verification=8Fq…"
A verification token. It proves nothing to an attacker except which products you use — and knowing you run Google Workspace is enough to make a fake login page convincing.
-
old-booking.baharicourt.example. IN CNAME app.retired-vendor.example.
A record left behind by a supplier you no longer use. If nobody holds that name at the vendor any more, someone else may be able to claim it and serve content on your subdomain.
The mail-related records get their own guide, because they are the ones that decide whether a criminal can email your guests or suppliers as you, and because the fix is free — SPF, DKIM and DMARC, explained properly.
The last line in that block is the one worth dwelling on. Stray records outlive the
suppliers that created them. A CNAME pointing at a service you cancelled is not
merely untidy: if the target can be re-registered by somebody else, they get to publish
content on a subdomain of your business, with your name on it. Cleaning up old records is
free and nobody schedules it.
What do certificate transparency logs reveal?
Every time you or a supplier obtains an HTTPS certificate from a publicly trusted authority — including the free automated ones — the hostnames on it are written into public logs, permanently. You can search them by domain. So can everyone else.
-
baharicourt.example, www.baharicourt.example issued 2026-05-02
Expected, and fine. This is the certificate a visitor sees.
-
pms.baharicourt.example issued 2026-04-19
Names the property-management system. An attacker now knows which vendor’s branding to copy in an email to your front desk.
-
vpn.baharicourt.example issued 2025-11-30
There is a remote-access gateway, and it is called what remote-access gateways are usually called. That is a target, and it is a target with a login page.
-
owa.baharicourt.example expired 2021-08-14
Expired five years ago and still in the log, because the log is append-only. It tells a stranger what you used to run, which is often still running somewhere.
This is not a flaw to be fixed; it is the mechanism working. The useful response is to be the first person to read your own log, not the last.
What does the domain registry reveal?
Registry data is published through RDAP, the protocol that replaced WHOIS. Personal contact details are usually redacted these days. The operational facts are not, and they are the ones that matter.
-
registrar: Example Registrar Ltd
The account that ultimately controls the domain. If it has no multi-factor authentication on it, nothing else you do to your DNS matters very much.
-
created: 2014-03-08 expires: 2026-08-14
Twenty days to expiry at the time of reading. A domain that lapses takes the website and the email with it, and re-registration is a race you may lose.
-
status: ok
"ok" means no restrictions are set. The status you want to see is clientTransferProhibited — the registrar lock that stops a transfer being initiated without you.
-
nameservers: ns1.serengeti-dns.example, ns2.serengeti-dns.example
Confirms the DNS provider independently of the DNS records themselves.
-
dnssec: unsigned
Not urgent for most small businesses, but worth asking your provider about once everything else is in order.
Why it matters: three chains that start here
Individually these are facts. The damage comes from combining two or three of them, and each of the following needs no access to anything.
- MX plus no DMARC becomes invoice fraud. The MX record says you use a well-known mail provider. The absent DMARC policy says nobody has told receiving servers to reject mail that only claims to be you. An email goes to a supplier, from your address, with different bank details. Nothing was hacked.
- A certificate log entry becomes a convincing phishing page.
pms.baharicourt.examplenames the property-management vendor. The login page is copied, the email is themed to match, and it goes to three people at the front desk who use that system every day. The credential that comes back is genuine, so every check passes afterwards. - An expiry date plus no lock becomes a hijack. A domain twenty days from expiry with no transfer lock, on a registrar account whose contact address is published, is a queue of one. Losing the domain means losing the website, the email and the ability to prove to anyone that you are you.
What can I actually do about it?
You cannot withdraw any of this. DNS is public because it must be, certificate logs are append-only on purpose, and registry entries are the record of a registration. What you can do is change what they say, and know it before someone else does.
- Fix the mail records first. Free, DNS-only, and it closes the most commonly exploited path — the full method is here.
- Put multi-factor authentication on the registrar account and the DNS account. These two accounts control everything else on this page, and they are usually the least protected things a business owns.
- Turn on the registrar lock and auto-renew, and make sure the address the renewal notice goes to is one somebody still reads.
- Retire stray DNS records, and check any
CNAMEthat points at a service you no longer pay for. - Be deliberate about hostnames in public certificates. A wildcard certificate covers subdomains without naming each one, which keeps individual names out of the log — at the cost of one key covering more, so it is a trade rather than a free win. The stronger move is not to publish internal services on public names at all.
- Read your own certificate transparency log occasionally. A certificate you did not ask for is worth knowing about early, and so is one a supplier issued for a subdomain nobody told you about.
- Publish a security.txt file at
/.well-known/security.txt(RFC 9116) with an address that reaches a person. Someone who finds a problem with your systems currently has nowhere to send it, and the second-best outcome is that they give up.
Checking your own, in about ten minutes
All of this is public, so you can do it yourself, right now, without any tool we sell.
-
dig A baharicourt.example +short ; dig MX baharicourt.example +short
Where the site and the mail live.
-
dig NS baharicourt.example +short ; dig TXT baharicourt.example +short
Who runs your DNS, and what your TXT records are announcing about the products you use.
-
dig TXT _dmarc.baharicourt.example +short
The one that decides whether a stranger can send email as you. An empty answer is a finding.
-
Search a certificate transparency log for %.baharicourt.example
crt.sh is the usual public search. Read the list of names as a stranger would.
-
Look the domain up at lookup.icann.org, or ask your registrar for the RDAP record
Check the expiry date, the status flags and whether the transfer lock is on.
If you would rather someone read it back to you in plain language, that is precisely what our free snapshot is: a domain and a work email in, a written report out, in one to two business days.
Is it legal for anyone to look at this?
Kenya's Computer Misuse and Cybercrimes Act 2018, as amended in 2025, draws the line in a place worth understanding — because it is the same line we work inside. Reading information that is already published infringes no security measure and is lawful. Connecting to somebody else's system to test it is a different act, and it requires their written authorisation.
Kenya has no responsible-disclosure safe harbour, and probing systems classed as protected — hospitals and payment systems among them — carries penalties of up to KES 25 million and 20 years. That is why a stranger reading your public records is doing something ordinary, and why anyone offering to "test" your systems without a signed authorisation is offering you a problem rather than a service.
If you find something with wider public consequences, the route in Kenya is the national response team, KE-CIRT, rather than a public post.
What we do with it, and what we do not
This is the material Reconesys works from. We read public records — DNS, mail records, certificate transparency logs and the domain registry. We never connect to your website or servers, never log in, and never test your defences.
One honest qualification, because the absolute version of that sentence would be false. Looking up a domain asks whichever name servers are authoritative for it, and those may be machines you or your provider run. So we will not tell you that nothing of ours ever reaches anything of yours. What is true is narrower and still strong: the only contact is a public DNS lookup of the same kind every mail server on the internet makes to find you, and nothing we do opens a connection to your website, your ports or your applications. Anyone claiming their outside-in check reaches nothing of yours at all is describing something that cannot be done.
Checks that would require connecting to your systems — what your live website actually serves, how your TLS is configured — stay switched off until you prove you control the domain or give us written authorisation. That is not a policy we could quietly change; it is enforced in our software.
The free snapshot is one domain, one report, written by a person and sent in one to two business days, with a rating, the three issues that matter most and a 30-day plan. It is not a penetration test. It is not monitoring — it is a picture of one day, and nothing will alert you if something changes afterwards. And it does not make your business compliant with the Data Protection Act.
- Read the worked sample report and judge it before giving us anything.
- Exactly what the snapshot covers — what it does not, at the same length, and why a person reads every report before it is sent.
- What the fix sprints cost, if you would rather the work were done for you than done by you.
- Where this sits in the Data Protection Act, if you are being asked for evidence.
Questions
Questions people ask about this
- Can someone find out what systems my business runs without hacking me?
- Largely, yes. Your DNS records name your web host, your mail provider and often the software-as-a-service products you have verified ownership with. Certificate transparency logs list the hostnames you have obtained HTTPS certificates for, which frequently include internal-sounding names like a VPN gateway or a booking system. The domain registry publishes your registrar, your renewal date and whether the domain is locked against transfer. All of it is published by design and none of it requires touching your systems.
- What is certificate transparency, and why are my server names in it?
- Certificate transparency is a set of public, append-only logs that publicly trusted certificate authorities write every certificate they issue into, so that a mis-issued certificate for your domain cannot stay hidden. Browsers expect to see proof of logging before they trust a certificate. It is a genuine security improvement, and the side effect is that every hostname you have ever requested a public certificate for is searchable by anyone.
- Can I remove a hostname from a certificate transparency log?
- No. The logs are append-only by design — that is the property that makes them useful, and entries cannot be deleted or edited. What you can do is stop adding new ones: use a wildcard certificate where it suits you, keep internal-only services off publicly trusted certificates, and retire services whose names are already published.
- Is it legal for someone to look up my DNS records in Kenya?
- Reading information that is already published — DNS records, certificate transparency logs, the domain registry — infringes no security measure and is lawful. Connecting to someone else’s systems to test them is a different act and requires their written authorisation under the Computer Misuse and Cybercrimes Act 2018, as amended in 2025. Kenya has no responsible-disclosure safe harbour, and probing systems classed as protected, which include hospitals and payment systems, carries penalties of up to KES 25 million and 20 years.
- What is the single most valuable thing to fix from this list?
- Your mail records. Whether a criminal can send convincing email as your business depends on three public DNS records — SPF, DKIM and DMARC — the fix is free and DNS-only, and invoice redirection by email is the fraud that most often takes money from Kenyan SMEs of this size.
Sources
Every legal and statistical claim above traces to one of these. We link to the publisher's own copy rather than to a summary, and name the provision in full so the citation survives a broken link.
Sources
- RFC 6962 — Certificate Transparency — IETF
The original design: public, append-only logs of issued certificates.
- RFC 9162 — Certificate Transparency Version 2.0 — IETF
The current revision of the same mechanism.
- Certificate Transparency — certificate.transparency.dev
Project documentation, including how to search the logs for your own domain.
- RFC 9082 — Registration Data Access Protocol (RDAP) Query Format — IETF
How a registry lookup is addressed. RFC 9083 defines the JSON that comes back.
- Registration Data Access Protocol (RDAP) — ICANN
What replaced WHOIS, and the lookup service at lookup.icann.org. For a .ke domain, the registry entry is held by KeNIC and your registrar.
- RFC 9116 — A File Format to Aid in Security Vulnerability Disclosure — IETF
The security.txt file: how someone who finds a problem can tell you about it.
- Computer Misuse and Cybercrimes Act, 2018 (Cap. 79C) — Kenya Law
The line between reading published information and connecting to someone else’s systems, and the penalties attaching to protected computer systems.
- National KE-CIRT/CC — Communications Authority of Kenya
Kenya’s national incident response team — the quarterly threat figures cited above, and the route for reporting something serious.