stride-analysis-patterns

STRIDE threat modeling analyzes each system component against six categories: Spoofing, Tampering, Repudiation, Information Disclosure, Denial of Service, and Elevation of Privilege. Start with a data flow diagram, identify trust boundaries, then enumerate threats per element and score them by impact and likelihood.

STRIDE is a threat classification framework where each letter maps to a security property: Spoofing targets authentication, Tampering targets integrity, Repudiation targets auditability, Information Disclosure targets confidentiality, DoS targets availability, and Elevation of Privilege targets authorization.

Prioritize threats by computing a risk score from impact multiplied by likelihood, each rated on a 1-4 scale. Scores of 12 or above are Critical, 6-11 are High, 3-5 are Medium, and below 3 are Low, producing a ranked mitigation list.

Yes, STRIDE maps naturally to DFD elements: external entities face spoofing and repudiation threats, processes face all six categories, data stores face tampering and disclosure, and data flows face tampering, disclosure, and denial of service. Trust boundary crossings highlight the highest-risk flows.

STRIDE works best for design-level analysis but does not replace penetration testing, code review, or runtime monitoring. It also requires regular updates since threat models are living documents that drift as the architecture evolves.