Jan 10 · 6 min read
Developing modern applications can be highly complex due to several disparate services that communicate with each other. Developers need to be deliberate about how they include business logic in their code. This can become time-consuming, and developers are forced to tack on security as an afterthought. Inevitably, this lax approach to security gives attackers a chance to easily compromise critical workloads. In fact, a recent report found that a shocking two-thirds of organizations lack a basic API security strategy.
In the following, we look at API security practices and how they play a critical role in helping secure modern distributed applications.
Traditionally, Application Programming Interfaces (APIs) have been used to help one server communicate with another server. However, with the advent of microservices, APIs have become essential for many-to-many communication between servers.
APIs are everywhere, and they get the job done, at least when it comes to building microservice applications that are reliable and fault tolerant. However, APIs aren’t inherently secure. And, due to them often being accessed through public networks, they are prime targets for attacks.
The best APIs are often well documented, and this documentation is readily available online, for example, this guide to the Zoom API. Such documentation is essential for developers but attackers can use it to uncover loopholes and break into applications without detection.
In fact, this type of attach has become commonplace. A recent survey has demonstrated that nearly 40% of organizations reported that more than one-half of their applications are exposed to the internet or third-party services via APIs.
As applications grow in value to the end user so do they grow in complexity. Developers are pressured to increase productivity. Startups like Tabnine and Raycast have had impressive funding rounds recently, indicating how important developer productivity has become. With this pressure to perform, developers don’t have the time to test each API connection for vulnerabilities or perform periodical penetration testing to ensure that new attack surfaces are not being introduced.
This is why API Security strategies tend to fail. Most organizations don’t have enough knowledge, expertise or manpower to develop holistic API security strategies that factor in every possible risk.
In addition, an effective API Security strategy becomes increasingly difficult as applications scale. Without proper direction, reviewers won’t even know where to start, let alone finish reviewing the entire codebase in time. This is where teams need to have a carefully planned API Security Checklist to ensure they don’t miss any vital security checks.
As mentioned, security, like most things, is easier when you have a good plan in place. Developers and reviewers can perform checks on APIs at their level without compromising on due dates. An exhaustive checklist would cover all bases and help teams streamline their API security strategy.
The most important checks to cover as part of this checklist are the OWASP API Security Top 10 threats. Both developers and reviewers should perform reviews to ensure that these threats are covered in the API security strategy. The top API security threats are well-documented, and teams can find plenty of material online to understand them and find ways to protect their APIs from these common threats.
Authorization and authentication checks are critical and should be performed rigorously to ensure APIs are fully secure. Teams can enter user input to see how APIs react and confirm no sensitive information is exposed on the client-side. Teams should also ensure any data filtration is done on the server-side and not on the client side. An authorized user might not see additional, sensitive data, but attackers can easily access this data. Rate limits and throttling are effective ways to prevent DDoS. If users have no limits on the number of times an API is called, attackers could deliberately ping that API till the server crashes.
APIs can also be compromised if the security configuration is incorrect. Security misconfigurations could happen if developers aren’t well-versed with API security processes or due to a compromised API update. Either way, these security flaws allow attackers to access critical workloads. Checks on security configuration are a vital part of any API security checklist.
Attackers can also manipulate APIs via hidden form fields. Security and development teams should use the inspect element option in web browsers to check for hidden form fields. If any hidden form fields are found, teams should enter random inputs to understand the API response and document the observations.
Injections can be a little tricky to deal with. Attackers can use SQL and NoSQL injections to an API to access data on the server. With command injections, attackers could try to manipulate the servers directly. To ensure attackers cannot access the server, teams can append OS commands in the API inputs and check the servers’ response.
Data encryption is another crucial step developers can take to ensure the security of their APIs. Teams should show relevant data only to authorized users upon decryption on the client-side. This ensures data isn’t accessible to malicious actors. Developers should architect an encryption strategy that is up to industry standards. Developers should also effectively abstract any data like passwords or keys so unauthorized personnel do not have access to sensitive data.
In the world of DevSecOps, security cannot be an afterthought. The security of an application is as important as the build. Organizations should develop proper ownership flow to ensure security is imbued in every deployment. If there is no clear ownership specifying who is in charge of what, it’s hard to collaborate, and in the end, the system gets compromised.
Tasks like encryption and elimination of sensitive data should be developers’ responsibility. At the same time, Ops teams should focus on updating APIs, eliminating obsolete APIs, and security configuration checks. Testing teams can own tasks like input fuzzing, parameter tampering checks, and authentication checks. It finally boils down to how teams are structured and how the responsibility can be shared across different members without creating silos.
An API security checklist evolves and changes over time as teams and their applications mature. Teams should use this checklist throughout the software development lifecycle. Teams should not put off security until the very end. They need to build secure applications from the start constantly. If APIs are built securely, it eliminates repeat work and security patches. Reviewers should ensure that the APIs pass every test on the checklist before an application is finally delivered. All the members involved in the projects, from developers, operators, testers, and security teams, should use this checklist to ensure proper collaboration, and to ensure API security doesn’t become a daunting process.
API security is multi-faceted and requires a lot of attention to detail. Today, security cannot be an afterthought. With thousands of API connections being used in workloads, it is hard for security teams to do patch-up work once a build is done. With a complete checklist, organizations can standardize API security across teams. And, teams can focus on what really matters while also making security a priority. To help with this, we created an API security checklist with all vital criteria listed in an organized manner. Download this checklist to plug any loopholes in your API security strategy. This checklist will also help teams truly collaborate and deliver products that aren’t just up to the mark, functionally, but are also secure.
Developing modern applications can be highly complex due to several disparate services…
On December 9th, the Log4Shell vulnerability (CVE-2021-44228) was published in a GitHub repository and made public….
While the big benefit of open source is the large developer community…