About TLCTC · 8 stops

From checkbox king to ten clusters

Who I am is not the topic here. The journey is. Pick your way to travel it.

No. 00 · The Journey · 8:16
1 of 8

Where I started

Confession first: I was a checkbox king.

My background was security engineering. Then I took on the CISO function, and the checkboxes appeared: ISO 27002, controls in, ticks out, audit green.

And next year, the same boulder up the same hill. That was the start of my Sisyphus story. I could name a control for almost anything. What I couldn't name was the threat each control was actually holding back.

2 of 8

Then I visited risk management

Risk management promised the missing half: cause, event, consequence.

The bow-tie is a beautiful model. Then I read the risk registers. Ransomware, data loss and missing patches all sat in one column, labelled “threat”.

Causes, outcomes and control gaps in a single list. You can't aggregate that, and you can't compare it across two organisations.

The cyber bow-tie
3 of 8

Due diligence

Surely someone had solved this already. I went looking.

Anything with cyber in its name or mission landed on my desk: standards, regulators, agencies, frameworks, threat reports. Months of reading.

Excellent catalogues of controls, techniques and outcomes. Not one shared, cause-based answer to the simplest question in the field: what is a cyber threat?

The gap analysis

Click a name to see where it leaves the gap.

4 of 8

Frustrated, because I felt I knew the answer

Yeah, we use the same damn words,
but we ain't speakin' the same song.

Cyber was a buzzword jungle: APT, zero trust, cyber resilience, threat landscape, next-gen everything. Even NIST's own glossary carries more than twenty definitions of “threat”, and they don't agree. One calls it a circumstance, another a likelihood, a third a cause of loss.

That frustrated me, because I had a hunch. An asset can only be weak in so many ways. If the weaknesses are finite, the threats that exploit them should be too. I couldn't prove it yet.

5 of 8

The thought experiment

Christmas 2022. Two weeks at the drawing board.

Axioms first: threats are causes, not outcomes. One generic vulnerability, one cluster. The attacker's identity never decides the class.

Then I walked through every component and interaction of an IT system and asked which generic vulnerability it carries. The hunch held: the list was finite, and ten clusters fell out. The original write-up is still on barnes.ch.

The original on barnes.ch Logical foundations
6 of 8

“So what?”

I took the ten clusters to my peers. They shrugged.

“Just another list.” “Which controls does it give me?” I recognised that question. It was the checkbox king's question, and I used to ask it myself.

Control catalogue fixation isn't a personal failing. Regulators ask for evidence of controls, not evidence of threat understanding, and a profession learns to answer what it's asked. Logic alone wasn't going to win. I had to show that naming the cause correctly changes the outcome.

7 of 8

So I built the application

Proof by use: if the ten clusters are right, they should carry real work.

The application paper takes the clusters into strategy and control planning. Its centrepiece is the control matrix: ten clusters by six CSF functions, every cell planned twice, as a local and as an umbrella control.

For operations, the attack path notation writes an incident as a sequence with velocity, boundaries and data risk events, and TLCTC+ extends it for national reporting. Around that grew the tools, the framework mappings and a blog that tests every claim on a real incident.

Application paper Tools Blog Attack paths

Ten clusters by six CSF 2.0 functions. Every cell is realised at two scopes, and an umbrella control only covers what sits under it.

Open the control matrix tool
8 of 8

A one-man show, with a tireless partner

TLCTC is a one-man show. It could never have reached this level without AI.

The axioms, the clusters and every classification ruling are mine, and so are the mistakes. But no one person writes two papers, fifteen tools, a dozen framework mappings, 165 posts and a set of films alone. AI was my co-writer, coder, sparring partner and harshest reviewer. It made a one-man show work like a team.

Shifting a profession takes more than one person: standards bodies, regulators, NIST, MITRE, and you.

Read the core paper Contribute on GitHub
1+AI
Bernhard Kreinz
Creator of TLCTC

Kreinz, B. (2026). A Cause-Oriented Cyber Threat Taxonomy: The Top Level Cyber Threat Clusters Framework. Zenodo. doi.org/10.5281/zenodo.20633176