Penetration testing for IoT products, from device to cloud

We run penetration tests on IoT products from the device to the cloud: firmware and the update path, the MQTT broker, backend and APIs, Kubernetes and cloud accounts, and the supply chain behind your containers. It is for makers of connected devices and their software teams. You get a report with prioritised findings and proposed fixes.

Request an intro call

Typical first stepTypical entry point: an intro call about the product, your timeline and what the test needs to answer. After that you get a written proposal.

Where teams usually start

  • The launch date is set. The device goes on sale soon, and nobody has attacked it from the outside yet.
  • A customer wants a security test. Their questionnaire asks when your product was last tested and how you handle findings.
  • Someone has reported a vulnerability. You want to know whether there are more before the next update ships.
  • New requirements for manufacturers. The Cyber Resilience Act sets requirements for connected products, for example on vulnerability handling and security updates.

What we test

  • Device and firmware: exposed interfaces, stored credentials and keys, the firmware itself.
  • Update path: whether the device accepts only signed firmware, and whether an older version can be pushed back onto it.
  • MQTT and broker: who may read and write which topics, and whether one device can reach another device’s data.
  • Backend and APIs: sign-in, permissions and the separation between customers.
  • Kubernetes and cloud: cluster configuration, permissions, network rules, secrets handling, on AWS, Google Cloud, Azure or your own hardware.
  • Supply chain: container images and dependencies checked for known vulnerabilities, and how new versions get into your pipeline.

How we work

  1. Scope.

    Together we set which devices, firmware versions, environments and accounts are part of the test and which questions it should answer. Scope, time window and your written authorisation are set down before we start. You get a written description of the test scope.

  2. Testing.

    We start with a threat model: who would attack what, and how. Then we test by hand and with tools, on the device, on the network and in the cloud, and record every finding with evidence.

  3. Report.

    Each finding with its severity, how we found it, what it allows and how to fix it, plus a summary for management. We go through the report with your team.

  4. Re-test.

    Once you have fixed them, we check the resolved findings again and record the result in the report.

What we run ourselves

What we test for, we put into practice in our own operations first:

Fixing what a test finds: Cloud-native DevOps & SRE. Building a platform hardened from the start: Kubernetes platforms, built and run.

Stack

  • Since 2 Oct 2026, the storage and system disks of our virtualisation hosts have been encrypted.
  • Every change to our platform is a commit, so it can be traced. Secrets sit in the repository, encrypted with SOPS and age.
  • In CI we scan every container image for known vulnerabilities and produce a software bill of materials (SBOM) for it.
  • In our product Fortlauf, the database itself enforces the separation between customers, through row-level security.
  • We take reports about vulnerabilities in our own services through our Security page.

Case studies

Connected-device vendor

Case study: full-stack security audit of an IoT product

For a maker of connected devices we ran a security audit from the cloud backend to the physical device. We found a critical MQTT vulnerability that could have allowed a device takeover and helped fix it through a secure firmware update process.

Our own product

Case study: running a SaaS on our own infrastructure

Fortlauf is our product. We built it, and we run it on our own hardware. In production since July 2026: tenant isolation in the database, immutable document storage, delivery by GitOps, measured recovery.

Who you will talk to

One fixed contact from the intro call to the report walkthrough.

Send a message

What we do not do

  • An assessment of whether NIS2 or the Cyber Resilience Act applies to your company. Whether and how NIS2 applies to you is a question for your legal advisers; we implement the technical measures.
  • No seal of approval and no statement that a product is secure. The report describes what we found within the agreed scope.
  • Fixing the findings is not part of the test. That is a separate project, with us or without us.

Frequently asked questions

What do you need from us before the test?

A description of the product and where it runs, test devices on the current firmware, access to the agreed environments (for example test accounts and a test instance of the backend), and a contact person on your team.

Do you test Kubernetes and cloud as well?

Yes, as a Kubernetes security assessment: we review your cluster configuration, permissions, network rules, how secrets are handled and the images running in it, on AWS, Google Cloud, Azure or your own hardware. How we build platforms ourselves: Kubernetes platforms, built and run.

What is in the report?

Each finding with its severity, the way we found it, what it could lead to and a proposed fix. On top of that, a summary for management and a list sorted by urgency.

Does the test help with requirements from the Cyber Resilience Act?

The Cyber Resilience Act sets requirements for makers of connected products, among them vulnerability handling and security updates. A penetration test gives you findings and measures you can implement technically to that end. Whether and how the Cyber Resilience Act applies to your product is a question for your legal advisers; we implement the technical measures.

Insights

Want your product tested?

Tell us what needs testing and by when you need the result.