Explore our new Cybersecurity Terms Glossary, explained by our experts.
Cybersecurity Terms Glossary
STRIDE gives a design team six prompts—spoofing, tampering, repudiation, information disclosure, denial of service, and elevation of privilege—for finding threats in a system.

What Is the STRIDE Threat Model?
STRIDE is a mnemonic for six categories of threat, invented by Loren Kohnfelder and Praerit Garg at Microsoft in 1999. Its value is that it converts a vague instruction—“think about what could go wrong”—into six specific questions asked of each part of a design. Teams who have never done security work before find threats with it, which is the whole point.
Each letter is the negation of a security property you want to hold.
| Threat | Property violated | Typical example | Usual mitigation |
|---|---|---|---|
| Spoofing | Authentication | Replaying a captured session token to act as another user | Strong authentication, short-lived tokens, mutual TLS |
| Tampering | Integrity | Altering a price or a quantity in a client-side request | Server-side validation, signing, integrity checks |
| Repudiation | Non-repudiation | A user denying a transfer that nothing recorded | Tamper-evident logging, digital signatures, synchronised time |
| Information disclosure | Confidentiality | An API returning fields the caller should never see | Encryption, output filtering, least privilege |
| Denial of service | Availability | A single query that exhausts the database connection pool | Rate limiting, quotas, timeouts, capacity planning |
| Elevation of privilege | Authorisation | A standard user reaching an administrative endpoint | Centralised authorisation checks, deny by default |
STRIDE works on a picture, usually a data flow diagram with four element types: external entities, processes, data stores, and data flows, separated by trust boundaries. The original per-element approach says that not every threat applies to every element, which narrows the work considerably:
A later variant, STRIDE-per-interaction, examines each source-destination pair instead. It finds more and takes longer; most teams start per element and move on if they need the depth. Either way, the elements that touch a trust boundary deserve the time, because that is where unvalidated input arrives.
Consider a file upload feature: a browser posts a document to an API, which writes it to object storage and records metadata in a database. Running STRIDE across it produces concrete findings in a few minutes.
Six questions, six mitigations to write into the backlog, none of which required a specialist.
STRIDE is memorable, needs no licence or tool, and produces a shared vocabulary between developers and security staff. It also maps cleanly onto the CIA triad extended with authentication, authorisation, and non-repudiation, which makes it easy to explain to auditors.
Its limits are real. It identifies threats but does not rank them, so pair it with CVSS, a likelihood-and-impact matrix, or the older DREAD scheme. It can generate long lists that overwhelm a small team if the whole platform is modelled at once. It says nothing about business logic abuse—a discount applied a thousand times is not obviously any of the six letters—and it was designed before cloud identity, machine learning, and supply chain compromise became primary concerns, so modern sessions often add categories for those. OWASP Threat Dragon and the code-first pytm project both support STRIDE if you want tooling, though a whiteboard and an hour will do for a first pass.
Use it as the generation step inside a wider threat modelling practice rather than as the whole of it. Our infrastructure and architecture review services run STRIDE sessions with delivery teams and leave the mitigations in your backlog, not in a report.
Spoofing, tampering, repudiation, information disclosure, denial of service, and elevation of privilege. Each category is the inverse of a property you want the system to have: authentication, integrity, non-repudiation, confidentiality, availability, and authorisation. Working through the six against each part of a design is a reliable way to surface threats without specialist experience.
Not strictly, but some picture of the system helps enormously. A data flow diagram showing external entities, processes, data stores, flows, and trust boundaries tells you which threats apply where, and stops the discussion drifting. A rough whiteboard sketch the team agrees on is worth more than a polished diagram nobody has checked.
STRIDE enumerates rather than ranks, so add a scoring step. For threats against released software, CVSS gives comparable scores. For design-stage threats, a simple likelihood-and-impact matrix, or a documented DREAD rubric, is usually enough. Whichever you choose, convert the output into backlog items with owners, otherwise the list is just a record of a meeting.
It remains a useful starting point, because identity, data flow, and authorisation threats are the same wherever the workload runs. It does not cover everything, though: business logic abuse, model poisoning, prompt injection, and supply chain compromise sit awkwardly in the six categories. Many teams run STRIDE first, then add domain-specific threat lists such as the OWASP guidance for large language model applications.
You May Be Interested In These
We use a few cookies that are needed to make this website work, and they are always on. We would also like to set analytics cookies, to see which pages are useful; functional cookies, so that our live chat works; and advertising cookies, to measure our campaigns. We will not set any of those until you say yes, and you can change your mind at any time from the Cookie Settings link in the footer.
Choose the cookies you are happy for us to use. Strictly necessary cookies are always on; everything else is off until you turn it on.
These cookies keep the website working: they hold your security checks, your form submissions and the choice you make here. The site cannot run without them, so they are always on.
These let our live chat open and remember your conversation, and let the newsletter sign-up work. Switch them off and the chat window will not appear.
These tell us which pages are read and how people find us, so that we can improve the site. The reports are about groups of visitors, never about you personally.
These let us measure our advertising and show you our adverts on other websites, such as LinkedIn, Google and Meta. Switch them off and you will still see adverts, but they will be less relevant.