MODULE_DEPLOYED // THREAT-ACTOR-PROFILE-OSINT

BUILDING A THREAT ACTOR PROFILE FROM OPEN SOURCES

Build a one-to-two page profile from public advisories and ATT&CK, grade every claim, and stop before fake personas, criminal forums, or bought leaks.

INTERMEDIATE

EST_TIME: 25 MIN

PHASES: 4

THIS_VISIT0 / 4 MARKED (0%)

Ticks last until you refresh. This page does not keep a score.

MISSION_PHASES
1
WHY PROFILE ANYONEDURATION: 5 min
2
SOURCES AND THE ADMIRALTY CODEDURATION: 7 min
3
ATT&CK, THE DIAMOND, AND THE PAGEDURATION: 8 min
4
PITFALLS AND A QUIZDURATION: 5 min
PHASE_001 // CURRENT_OBJECTIVE
WHY PROFILE ANYONE
NODE: 1 / 4

A shrug is not a threat model

"Hackers" is not a threat model. A ransomware crew going after Australian healthcare, a state-backed group interested in mining and energy, and a teenager selling stolen logins all behave differently and need different defences. A profile answers three questions: who is likely to target us, what do they want, and how do they usually operate? That lets you prioritise. Without one, you are defending against everything, which in practice means nothing in particular.

The frames for this lesson are MITRE ATT&CK (Groups and Navigator), the Admiralty Code for grading sources, and the Diamond Model for a single intrusion. The product is one or two pages someone can use.

Types of actor, briefly

  • Cybercriminals: money. Ransomware, business-email compromise, credential theft for resale.
  • State-sponsored: espionage, pre-positioning, occasionally disruption.
  • Hacktivists: publicity and causes. Defacements, denial of service, leaks.
  • Insiders: access they already have, with mixed motives.
"Hackers" is a shrug with a budget attached. A profile is what you write when the shrug stops being good enough.
ASK_A_QUESTION