Recent Posts
Archives

Posts Tagged ‘STRIDE’

PostHeaderIcon [DevoxxFR2026] Threat Modeling: Collaborative Security from Design to Deployment

Lecturer

Barbara Teruggi is a Security Architect with extensive experience in threat modeling and secure system design. Her work emphasizes practical, inclusive approaches to embedding security throughout the development lifecycle.

Abstract

Threat modeling has evolved from a specialized security practice to an essential component of modern product development. As cyber threats intensify and delivery pressures increase, security risks often surface too late—during QA or post-deployment. Barbara Teruggi demonstrates that developers, who best understand application behavior, must participate actively from the outset. Through the four-question framework, STRIDE categorization, data flow diagrams, and risk analysis, she shows how cross-functional teams can identify and mitigate threats collaboratively. Using a “digital house” analogy, the presentation makes complex security concepts accessible and actionable for all participants.

Understanding the Need for Proactive Security in Development

Modern software faces relentless attacks exploiting design flaws rather than implementation bugs. The OWASP Top 10 consistently highlights insecure design as a leading cause of breaches. Traditional approaches—adding security after development—prove costly and ineffective. Threat modeling shifts security left, integrating it into the design phase where changes remain feasible and economical.

The analogy of building a house illustrates the concept perfectly. Moving into a seemingly perfect property only to discover structural problems creates expensive headaches. Similarly, deploying applications with latent vulnerabilities leads to emergency fixes or worse, breaches. Threat modeling acts as the architectural review, identifying weaknesses before construction begins.

Core Concepts: Threats, Vulnerabilities, and Risk

Clear definitions ground effective threat modeling. A threat represents a potential event exploiting weaknesses to cause harm. A vulnerability constitutes the exploitable flaw itself. Risk combines likelihood and impact, measuring potential damage.

The STRIDE framework categorizes threats systematically: Spoofing, Tampering, Repudiation, Information Disclosure, Denial of Service, and Elevation of Privilege. This structured approach ensures comprehensive coverage of potential attack vectors.

Data Flow Diagrams (DFDs) provide visual representations of system architecture, highlighting trust boundaries and data movement. These diagrams prove invaluable for identifying where security controls must be enforced.

The Four-Question Framework for Threat Modeling

Adam Shostack’s methodology organizes threat modeling into four practical questions. “What are we working on?” establishes scope through architecture diagrams, technologies, data flows, and compliance requirements. This phase compiles all relevant security context.

“What can go wrong?” involves enumerating threats using STRIDE or similar taxonomies. Teams brainstorm potential attack scenarios against system components.

“What are we going to do about it?” focuses on risk analysis and treatment. Prioritization considers business impact, likelihood, and mitigation costs. Strategies include avoidance, acceptance, transfer, or mitigation.

“Did we do a good job?” validates the process through testing, documentation updates, and retrospective analysis. This ensures continuous improvement and knowledge capture.

Collaborative Participation and Practical Implementation

Effective threat modeling requires diverse perspectives. Security professionals provide expertise, but developers understand implementation details, architects grasp system structure, product owners represent business needs, and UX designers ensure user experience alignment. This cross-functional involvement produces more robust outcomes than siloed security reviews.

The RACI model (Responsible, Accountable, Consulted, Informed) clarifies roles throughout the process. Regular checkpoints maintain alignment and prevent knowledge gaps.

Implementation need not be overwhelming. Start small with existing applications or new features. Leverage existing diagrams and documentation. Iterate gradually, building institutional knowledge and patterns over time.

Conclusion

Threat modeling transforms security from reactive firefighting to proactive design practice. By involving all stakeholders, using structured methodologies, and maintaining living documentation, teams build more resilient systems. The collaborative nature makes security everyone’s responsibility, ultimately producing better products with fewer “oops” moments. As threats evolve, organizations embracing threat modeling gain significant advantages in security posture and development efficiency.

Links: