STRIDE per Interaction Analysis
What you'll learn: How to examine each interaction between system components for security properties and map any violations to specific STRIDE threats.
Why Interactions Matter
You've learned STRIDE per Element, which examines components themselves. Now we shift focus to the connections between components—the arrows on your DFD. Each interaction (like a user logging in, or a database query) has security properties that can be violated.
The Core Approach
For every interaction (data flow) in your system, ask about these security properties:
- Authentication: Is the identity verified?
- Integrity: Can the data be tampered with in transit?
- Non-repudiation: Can actions be denied later?
- Confidentiality: Is the data protected from eavesdropping?
- Availability: Can the flow be blocked or flooded?
- Authorization: Is privilege checked at both ends?
When a property is violated or weak, map it to a STRIDE threat:
- No authentication → Spoofing
- No integrity checks → Tampering
- No logging → Repudiation
- No encryption → Information Disclosure
- No rate limiting → Denial of Service
- Insufficient authorization → Elevation of Privilege
Analogy
Think of a postal delivery route. The interaction is the letter traveling from sender to recipient. You analyze: Is the sender verified? (authentication) Is the envelope sealed? (integrity) Can it be intercepted? (confidentiality) Can someone flood the mailbox? (availability)
Practical Example
Interaction: Web browser → Web server (HTTP request)
- Confidentiality violated (no encryption) → Information Disclosure
- Integrity violated (no HTTPS) → Tampering
- No rate limiting → Denial of Service
Key Takeaway: STRIDE per Interaction analyzes data flows between components by checking which security properties each interaction lacks, then mapping those gaps to specific STRIDE threat categories.