Personal Assistants like Alexa, Cortana or Siri have a lot of applications; however, many require special APIs to solve the posed problems. This Bachelor Thesis formally defines the problems FULLMATCH and FLEXMATCH, which require coordination of services on a temporal level. The developed API, named â€™TempCoâ€™, solves these problems and its components are described and explained here in detail. It leverages the Semantic Web through JSON-LD and can parse data from various sources by using the Schema.org vocabulary. All algorithms were written in TypeScript which allows the code to be portable and run on many platforms. A standalone Node.js HTTP REST server implementation is also provided that listens to POST requests, enabling algorithm usage through the network. We analyse all algorithms on Performance to identify practical limits and runtime complexities. Results show that parsing JSON-LD objects can be slow for real time environments and the FLEXMATCH algorithm best works for small or constrained inputs. In contrast, the FULLMATCH algorithm is very scalable and can solve typical problem instances in a matter of seconds. Finally, we describe possible use cases, showing the versatility and applications of both algorithms to enhance Personal Assistants.