I did some research and found there is no particular format that Atom feed sticks to. There is no specific way to parse it. There are different architecture models for implement Atom feed generators. They can be implemented in push or pull models. If it is pull model, we can use Connect-HTTP to read a feed.
Generally, Atom/RSS feed is designed to deliver XHTML content, but there is no restriction on the format. It can be simple HTML or a complex XHTML document or even well-formed XML. Parsing needs depend on the message structure.
Pega does not have any standard way to parse Atom feeds as of 7.2.2. Feeds are fairly flexible for passing display content across systems. So, there is no limitation on data structure to pass through this feed. So, even if Pega provides a standard way to send and receive feeds, the integration has to limit to connection protocol. Feed burning/consumption would still stay open. We cannot expect a standard parsing rule like XML, where there are specific guiding principles to structure the message.