Calling multiple functions using plugins
Semantic Kernel accomplishes AI orchestration by allowing you to integrate plugins into the kernel.
A Semantic Kernel plugin is a collection of one or more KernelFunction objects that can be discovered and invoked by the kernel in response to a specific input. This lets the kernel choose which function or functions might be relevant to responding to a specific prompt such as a message from a user.
Plugins can be defined by inheriting from the KernelPlugin class or via one of the simpler import methods built into kernel instances.
We’ll take the latter approach because we want to create a simple plugin that combines some of the various functions we’ve created so far:
KernelPlugin plugin =
kernel.ImportPluginFromFunctions("FootballPlugin",
"A collection of functions related to football",
functions: [coach, timeFunc, teamInfoFunc, searchFunc...
