1. Topology: Classify if the threat is a Bridge Cluster (#8, #9, #10) crossing a domain boundary, or an Internal Cluster (#1–#7) operating within a control regime.
2. Velocity (Δt): Measure the time between steps. VC-4 (seconds–ms), VC-3 (minutes), VC-2 (hours), VC-1 (days–months).
3. Boundary Notation: Use ||[context][@Source→@Target]|| for all bridge cluster boundary crossings. V2.1 adds transit (⇒) and intra-system (|[type]|) operators.
Visual Decision Flow
Decision Tree: Step by Step
“Ransomware” or “data breach” are outcomes, not clusters. Classify each step by the generic vulnerability exploited. One step = one cluster (Axiom VI).
Q1: Is this targeting Human Psychology? (Bridge)
Does it exploit trust, fear, or urgency to bypass technical controls?
Crosses from Human Domain to Cyber Domain.
Examples: Phishing, Vishing, MFA Fatigue (Psychological pressure).
Q2: Is Physical Access Required? (Bridge)
Does it require touching hardware or exploiting physical layer signals?
Crosses from Physical Domain to Cyber Domain.
Examples: USB insertion, Device Theft, TEMPEST.
Q3: Is a Trust Boundary Crossed? (Bridge)
Does the attack leverage implicit trust in a third-party component, update, or pipeline?
Crosses from Vendor Domain to Org Domain.
V2.0 Syntax: Use the ||[context][@Source→@Target]|| boundary operator.
#10 ||[dev][@Vendor→@Org]|| → #7
R-SUPPLY: #10 is placed at the Trust Acceptance Event (TAE) — the moment the trust artifact becomes authoritative inside the target domain.
Q4: Communication Interception? (Internal)
Is the attacker positioned between communicating parties within a channel?
Operates within the communication domain.
Q5: Resource Exhaustion? (Internal)
Is the attacker overwhelming a resource (bandwidth, CPU, memory, connections) to deny availability?
Operates within the target domain through resource exhaustion.
Examples: DDoS, SYN flood, application-layer flooding, amplification attacks.
Q6: Credentials? (R-CRED & Axiom X)
Credential Duality — Acquisition vs. Use are always separate steps:
- ACQUISITION Phase: Map to the enabling cluster (e.g., #9 Phishing, #1 lsass dump, #2 SQLi). This is a Data Risk Event
+ [DRE: C]. - USE Phase: ALWAYS maps to #4 Identity Theft, regardless of acquisition method (R-CRED).
Q7: Foreign Executable Content (FEC)?
Is code executing that was not part of the original system design?
R-EXEC: If Foreign Executable Content executes, a #7 step with fec_executed: true MUST be recorded at the execution moment.
Includes Scripts, Binaries, and LOLBAS usage (where #1 invokes the tool, and #7 executes the payload).
Q8: Exploiting Implementation Flaws?
Does it require Exploit Code to trigger a bug (buffer overflow, injection)? R-ROLE: Classify by the role of the flawed component relative to the attacker.
Q9: Logic/Feature Abuse?
Abusing legitimate functionality without code flaws and without foreign code?
Examples: Parameter tampering, API abuse, Admin config changes.
Attack Patterns & Velocity
#9 ||[human][@External→@Org]|| →[Δt=24h] #7 →[Δt=5m] #4 →[Δt=15m] (#1 + #7) + [DRE: Ac]
(Phishing → Malware → Cred Use → (Admin Abuse + Encryption) → Data Inaccessible)
#1 →[Δt=weeks] #10 ||[dev][@Vendor→@Org]|| →[Δt=days] #7 + [DRE: C, I]
(Repo Abuse → Trust Acceptance Event → Malware Execution)
#4 →[Δt=1m] #1 →[Δt=5m] #9 ||[human][@External→@Org]|| → #4
(Cred Use → Abuse MFA Request → Fatigue (Social) → Token Use)
Notation Quick Reference
→sequential step(#X + #Y)parallel execution→[Δt=value]→attack velocity||[ctx][@Src→@Tgt]||boundary crossing (bridge clusters)
+ [DRE: C]Confidentiality+ [DRE: I]Integrity+ [DRE: A]Availability (general)+ [DRE: Av]data gone/unreachable+ [DRE: Ac]data present but unusable (e.g., ransomware)
V2.1 Boundary Extensions
These are additive and backward-compatible with V2.0 notation.
Marks spheres that carry/relay the attack but are not the source or target.
||[human][@Attacker⇒@SMSProvider→@Victim]||
R-TRANSIT-3: Vendor code running on the target device is NOT transit. It is the attack surface (classify by R-ROLE).
Marks boundary crossings within a single host or system. Single pipe delimiters.
#3 |[sandbox][@renderer→@os]|
4 types: sandbox, privilege, process, hypervisor. R-INTRA-7: These never change cluster classification — they are observability annotations only.
References
- Kreinz, B. Top Level Cyber Threat Clusters (TLCTC), White Paper V2.0
- TLCTC Enumeration V2.0 Documentation.