Agents are another common implementation in the Elixir concurrency world, especially around the context of state management. agents are pretty much designed for that sole purpose: they exist to be simple wrappers around state! They're incredibly simple to work with. Let's open an IEx window (if we don't already have one open) and begin implementing a simple agent:
iex(1)> {:ok, agent} = Agent.start_link(fn -> %{} end){:ok, #PID<0.174.0>}
We start off by initializing our agent with Agent.start_link/1. This function takes in a single argument, which is the function that returns the agent's initial state. In the preceding case we're starting off with a completely blank map:
iex(2)> Agent.update(agent, fn state -> ...
