Explore our new Cybersecurity Terms Glossary, explained by our experts.
Cybersecurity Terms Glossary
API security is the practice of protecting application programming interfaces from abuse, covering discovery, authentication, authorisation, input validation, rate limiting, and monitoring.

What Is API Security?
API security is the set of practices and controls that protect application programming interfaces—the endpoints that mobile apps, web front ends, partners, and internal services use to exchange data. Because an API exposes business logic and data directly, with no user interface in the way, a single missing check can let an attacker pull millions of records with a simple script. Securing APIs means knowing which ones you have, verifying who calls them, limiting what each caller can do, and watching how they are used.
Modern applications are thin front ends sitting on dozens or hundreds of APIs. Each endpoint is an entry point, and many are undocumented: old versions left running after a migration (“zombie” APIs), or endpoints built quickly for a single partner and never registered (“shadow” APIs). APIs also tend to return more data than the screen displays, trusting the client to filter it—an assumption attackers exploit by calling the endpoint directly with tools such as Burp Suite or Postman.
Traditional perimeter controls help less than you might expect. A web application firewall recognises an SQL injection payload, but a request for /api/v2/accounts/88213 made by the wrong customer looks entirely legitimate. Most serious API breaches exploit logic, not malformed input.
Three of the top five are authorisation failures, which is why API testing overlaps heavily with broken access control. Inventory management (API9) deserves equal attention: you cannot protect an endpoint you do not know exists.
In September 2022, the Australian telecoms provider Optus disclosed a breach affecting around 9.8 million current and former customers. Public reporting attributed it to an internet-facing API that did not require authentication and used predictable customer identifiers. In January 2023, T-Mobile in the US disclosed that an attacker had abused one of its APIs to obtain data relating to around 37 million accounts over several weeks. Neither case needed an exploit in the traditional sense—just requests the APIs were happy to answer.
Maintain a living inventory of every API, version, and environment, ideally generated from OpenAPI specifications and confirmed by traffic analysis. Retire old versions deliberately rather than leaving them running. External attack surface discovery helps catch forgotten hosts and test environments exposed to the internet.
Use established standards—OAuth 2.0 and OpenID Connect for users, mutual TLS or signed tokens for service-to-service traffic—and never invent your own scheme. Validate token signatures, audience, and expiry on every request. Then check, in server-side code, that the caller may access that specific object and that specific field. The UK Open Banking standard, built on the OpenID Foundation’s FAPI security profile, is a good example of how strict this can be made for high-value data.
An API gateway gives you a single place to enforce policy and collect logs. Feed those logs into your monitoring platform and alert on patterns such as one token enumerating sequential IDs, sudden spikes in error rates, or a client requesting far more records than normal.
Automated scanners can find misconfigurations and known vulnerabilities, but authorisation flaws require testers who understand the business rules and who test with multiple accounts and roles. A good assessment starts from the API specification, adds endpoints found in mobile apps and JavaScript bundles, and then attempts object, property, and function-level bypasses. Our web application security services include REST, SOAP, and GraphQL API testing performed this way.
No. A gateway is valuable for authentication, rate limiting, logging, and blocking obviously malicious traffic, but it usually cannot decide whether user A may read record B. That decision depends on business logic inside the application. Treat the gateway as one layer and keep object-level and function-level authorisation checks in the backend code.
Shadow APIs are endpoints running in production that the security team does not know about, often created for a quick integration or a test. Zombie APIs are deprecated versions that were never switched off after a newer version launched. Both tend to lack current security controls and monitoring, which makes them attractive targets for attackers.
GraphQL exposes a single endpoint with a flexible query language, so controls move from the URL level to the field and resolver level. Specific risks include introspection revealing the whole schema, deeply nested queries that exhaust server resources, and batching that bypasses rate limits. Authorisation must be enforced in each resolver, not only at the endpoint.
At least annually, and whenever you release significant new endpoints, change authentication, or onboard a major partner integration. For fast-moving teams, combine periodic manual testing with automated security tests in the CI/CD pipeline and continuous discovery, so new or changed endpoints are caught between formal assessments.
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.