Week 1 · D1🔒 LockedAbout 3 hours
Threat modeling: the lethal trifecta, trust boundaries, and how to use two OWASP lists as checklists
Start with the lethal trifecta test to see where an agent is actually dangerous: once private data, untrusted content and outbound communication coexist, one successful prompt injection becomes one data exfiltration. Then draw it as a trust boundary diagram and walk both OWASP lists against it.
Start with the lethal trifecta test to see where an agent is actually dangerous: once private data, untrusted content and outbound communication coexist, one successful prompt injection becomes one data exfiltration. Then draw it as a trust boundary diagram and walk both OWASP lists against it.
Interview questions
What is the lethal trifecta, and why is it a compositional risk rather than a single-point defect?什么是致命三件套?为什么说它描述的是组合风险而不是单点缺陷?
Common in ChinaCommon overseasBasic#threat-modeling#lethal-trifecta🔒 Unlock this chapter to see the answer keys.
What is the difference between direct and indirect prompt injection, and which is harder to defend against?直接注入和间接注入有什么区别?哪一种更难防,为什么?
Common in ChinaCommon overseasIntermediate#prompt-injection#attack-surface🔒 Unlock this chapter to see the answer keys.
Given an agent already in production, how would you map its trust boundary and find its highest-risk path in half an hour?给你一个已经上线的 Agent,你怎么在半小时内画出它的信任边界并找出最高危的那条链路?
Common in ChinaCommon overseasDeep dive#threat-modeling#trust-boundary🔒 Unlock this chapter to see the answer keys.
