Service · Security review

Application security review, from login to firewall.

We check your system the way an attacker would look at it: how people sign in, who can reach what, which doors the servers leave open, and where the keys are kept. Our application security review ends in a written list, the most serious first, in plain language. Then the same senior engineers fix what you choose, so the report doesn’t sit in a drawer.

A person reads every message. No pitch unless you ask for one.

an illustration of a review’s findings list: what was checked, what is open and what is fixed

01 · What goes wrong

It works, so it feels safe. Nobody has checked.

The gaps a review looks for first, the way an owner would describe them.

  1. The admin login is one password away.

    No second step at sign-in, a password the whole team shares, and no limit on wrong guesses. Whoever gets that one password runs the system.

    Exposed
  2. The database answers to the whole internet.

    It was opened once to make setup easier and never closed. Everything the app stores is one leaked password from being copied.

    Open
  3. Keys and passwords sit in the code.

    Payment, email and cloud keys were pasted in to get things working, and now live in every copy of the code and its history.

    Leaked
  4. The rules are only on the screen.

    The page hides the admin button, but the server never checks who is asking, so anyone who knows the address can do what an admin does.

    Unchecked
  5. Nothing has been updated since launch.

    The software underneath has known flaws with published fixes. The bar at the top of this site lists the ones attackers are using right now.

    Stale
  6. Nobody would know if someone got in.

    No record of who signed in or what changed, no alert when something looks wrong, and a backup nobody has ever tried to restore.

    Blind

02 · Under attack now

The software flaws attackers are using right now.

The newest entries on the US government’s list of flaws known to be used in real attacks. If your system runs any of this software, it needs its update.

  1. Nov 17Perforce Puppet Enterprise Deserialization of Untrusted DataCVE-2026-115731 (the public record, opens in a new tab)
  2. Nov 13Check Point Multiple Products Incorrect Permission Assignment for Critical ResourceCVE-2026-123039 (the public record, opens in a new tab)
  3. Nov 13UnrealIRCd Improper Neutralization of DelimitersCVE-2026-114713 (the public record, opens in a new tab)
  4. Nov 10Cisco BroadWorks Improper Restriction of XML External Entity ReferenceCVE-2026-20320 (the public record, opens in a new tab)
  5. Nov 10Google Chromium Use-After-FreeCVE-2026-125350 (the public record, opens in a new tab)
  6. Nov 10Microsoft Office Outlook Missing User Interaction for Potentially Unsafe ActionCVE-2026-124194 (the public record, opens in a new tab)
  7. Nov 10Barracuda Networks Email Security Gateway (ESG) Authentication Bypass Using an Alternate Path or ChannelCVE-2026-116606 (the public record, opens in a new tab)
  8. Nov 10Openwave Messaging Limited Webtop Media Transcoder Incorrect Resource Transfer Between SpheresCVE-2026-111793 (the public record, opens in a new tab)

Source: the Known Exploited Vulnerabilities catalog (opens in a new tab) published by CISA, the US cybersecurity agency. Each line is the day CISA added the flaw and its name in the catalog’s own words. Read on November 17, 2026, and read again twice a day. Synaptech is not affiliated with CISA.

Running any of this? Ask us to check your system

03 · What we do

Security review and hardening, from the sign-in page to the server.

We write down what we will look at before we look, then check each part and write down what we find.

  • 01Sign-in

    Logins and authentication

    A second step at sign-in where it matters, sign-in with the Google or Microsoft accounts your team already has, limits on wrong guesses, sessions that end, and a password reset that can’t be used to take over an account.

  • 02Access

    Who can reach what

    Roles and permissions checked on the server, not just hidden on the screen, admin actions that leave a record, and accounts closed when someone leaves.

  • 03Firewall

    Firewalls and the network

    Firewall rules that leave open only the doors the app needs, databases and admin tools taken off the public internet, and limits on how fast anyone can knock.

  • 04Backend

    The backend and its API

    Every request checked for who is asking and what they may see, input checked before it is used, uploads and payment callbacks verified, and errors that tell an attacker nothing.

  • 05Data

    Keys, data and backups

    Keys and passwords moved out of the code and replaced, data encrypted on the way and at rest, each customer’s data kept apart from the next, and a backup that has been restored once to prove it works.

  • 06Upkeep

    Updates, logs and alerts

    The software underneath brought up to date, a record of sign-ins and changes, alerts when something looks wrong, and a short written plan for the day something does.

We fix what we find. The engineers who review the system make the changes, test them, and show you what changed.

04 · Proof you can check

Security we built into systems that handle money and personal records.

Sign-in, access and data isolation in products we built, each credited to the client it was built for.

The work behind the numbers

The number

41 API modules in one operations platform, replacing four legacy apps

Automotive retail (multi-rooftop dealer group)

Four legacy apps became one platform for a 12-rooftop auto group.

Clienta 12-rooftop auto group
  • ClientA 12-rooftop auto groupRole-based access with per-employee grants and revokes. A permission override needs a reason and lands in a full audit trail, and the menu each person sees is served by the server, never trusted from the browser.
  • ClientWillowOne-time passcodes and phone verification at signup, separate test, development and production environments, and a database with row-level security and an append-only event ledger.
  • Venture we co-operateCravebleOne backend ran every brand’s store, with each store’s data isolated in the database itself by row-level security, and owner roles and per-store keys scoping access.
  • ClientSkillScheduleRole-based access across parent, coach and admin portals, with Apple and Google sign-in in the mobile app.

05 · How it starts

Agree what we look at, review it, fix it, keep it that way.

The same order every time. Nothing is tested against a live system without your written go-ahead.

We look only at what you agreed to in writing, with the least access the review needs, and a senior engineer reviews every change before it ships.

  1. Step 1: Tell us what you run

    Send a note with what the system is, where it runs and what worries you. A person reads it and writes back with what we would look at first.

    Out a written reply from a person

  2. Step 2: Agree the scope and the access

    We agree in writing what we will look at and how. Access is the least the review needs, read-only wherever that is enough, and it is removed when the work ends.

    Out a written scope and access list

  3. Step 3: The review

    We go through sign-in, access, the network, the backend, keys, updates and backups, and write down what we find in plain language, the most serious first.

    Out a written report, worst first

  4. Step 4: The fixes

    You choose what to fix. You get the scope, a timeline and an estimate in writing, then we make the changes, test them, and show you each one.

    Out the fixes, tested and written up

  5. Step 5: Keep it that way

    A short list for your team: how updates get applied, how accounts are opened and closed, and when to ask for another look, such as after a big change.

    Out a short upkeep list

06 · Straight answers

Questions owners ask before a security review.

What kinds of systems do you review?

Web apps, online stores, customer and staff portals, the backends of mobile apps, and the cloud accounts they run on. They are the kinds of systems we build ourselves, so we know where the gaps tend to be.

Do you need access to our systems?

Yes, and as little as the review needs. We agree it in writing first, ask for read-only access wherever that is enough, and remove it when the work ends.

Will the review take our site or app down?

It is planned so it doesn’t. Most of the review is reading how the system is built and set up. Anything that could affect a live system is agreed with you first and done at a quiet hour.

Is this a penetration test or a compliance audit?

No. It is an engineering review: we read how the system is built and set up, find the gaps, and fix them. It does not come with a certificate. If a customer or an insurer has asked you for a formal test or a certificate, tell us, and we will say plainly whether we are the right team for it.

What is the list in the bar at the top of the site?

It comes from the catalog of known exploited vulnerabilities published by CISA, the US cybersecurity agency: software flaws that attackers are known to be using. We read it twice a day. If your system runs any software named there, it needs its update, and we can check that for you.

What does a security review cost?

It depends on the size of the system, and the plan says so in writing: the scope, a timeline and an estimate before you commit. We don’t publish prices or packages, because no two systems need the same work.

07 · The front door

Find out what an attacker would find first.

Tell us what you run and what worries you. A person reads every message and writes back.

A person reads every message. No pitch unless you ask for one.

info@synaptech.io

Ask for a security review

For a reply by text, add a US mobile number and tick the box below.

We use your details to reply, and to text you only if you tick the box. Privacy policy

Ask for a reviewBook a call