An API (application programming interface)*[1]​ is a piece of code that allows two applications to communicate with each other to share information and functionality. They are generally used in programming libraries.[2]​.

For example, if you have a recipe app on a mobile phone and when you operate this app you perform a search, you can use an API so that this application communicates with the recipe website, requests the recipes that meet the search criteria, returns the results and later, after the user selects the desired recipe, requests and downloads it.

In this way, an API is a bridge to connect different applications and make them work more efficiently and effectively.[2]​.

An API represents the ability to communicate between software components. It is the set of calls to certain libraries that offer access to certain services from the processes and represents a method to achieve abstraction "Abstraction (object-oriented programming)") in programming, generally (although not necessarily) between the lower and higher levels or layers of the software. One of the main purposes of an API is to provide a set of general-purpose functions, for example, to draw windows or "Icon (Computing)" icons on the screen. In this way, programmers benefit from the advantages of APIs by making use of their functionality, avoiding the work of programming everything from the beginning. APIs are also abstract: the software that provides a certain API is usually called the implementation of that API.

For example, you can see the task of writing "Hello World" on the screen at different levels of abstraction "Abstraction (object-oriented programming)"):.

As you can see, the first option requires more steps, each of which is much more complicated than the steps in the following options. Furthermore, it is not at all practical to use the first approach to represent a large amount of information, such as an encyclopedic article on the screen, while the second approach simplifies the task by eliminating one step and making the rest simpler, and the third way simply requires typing "Hello World". However, high-level APIs generally lose flexibility; For example, it is much more difficult in a web browser to rotate text around a dot with a blinking outline than to program it at a low level. When choosing to use an API you must strike a certain balance between its power, simplicity, and loss of flexibility.