1

Is there a standard diagram format or language for illustrating a security model and what information is needed to access specific accounts? Something like UML for passwords and 2FA?

I was recently reading Sean Coonce's blog post on how he lost $100k in crypto to a SIM port hack.

Or here's another example where "Getting into Amazon let my hackers get into [an] Apple ID account, which helped them get into Gmail, which gave them access to Twitter".

Is there any kind of established systematic way of identifying these kinds of vulnerabilities?

In both of these examples the hackers were able to gain access to one system because they were able to gain access to other systems.

I'm envisioning a graph diagram something like this:

enter image description here

Where the nodes are either pieces of information, or systems to be accessed.

Does something like this already exist and is in use by security professionals?

  • 1
    Given that everyone's personal situation will be different, I doubt there's any predefined pattern, but this sort of threat analysis (if X is compromised, will Y be compromisable?) is done all the time. Now, if only I knew enough about it to write up an answer... – Fund Monica's Lawsuit Jul 6 '19 at 0:20
3

I am not aware of a diagram set but there is veris: veris framework

It translates the "Who did what to whom using which attack vector" in a structured JSON Format. It uses a 4 dimensional model for threats to describe any incident.

  1. Actors
  2. Actions
  3. Assets
  4. Attributes

So the example model from above would be slightly different, you can model every incident with those 4 dimensions and then link them together via the asset used and its attributes.

I personally like the structure when establishing kill chains for apt attacks.

Another option would be the attack tree model from Bruce Schneier Attack Tree Model

There you go backwards through the steps for a given attack, starting by the affected asset and then model every step the attacker has to take to gain access to the asset. This would model the above given scenario from the bank account backwards and create a path for the attacker. Then you can assign probabilites and cost to every node to model the most likely way any attacker would go.

A more sophisticated approach ist the SANS Threat modeling approach This creates a campaign out of the attack tree an then describes every step as an scenario.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.