To inspect plain-text contents of communications over SSL, interception proxies insert themselves in the flow of traffic and terminate the client's request. The interception proxy makes a second request on behalf of the client to the server. This behavior causes what was an end-to-end session to instead be two separate, but related, point-to-point sessions. The goal is to allow access to the plain-text session contents while transferring between the two encrypted sessions.
Interception proxies may be implemented in several ways, depending on their purpose and the type of inspection they perform. Perhaps the most common scenario is to integrate with an existing HTTP or SOCKS proxy. However, Deep Packet Inspection devices, next-generation firewalls, application-aware firewalls, and data loss prevention (DLP) systems are other possible points. Interception may also reside on the client itself as part of an endpoint protection suite, or even as a web-browser plugin.
Despite the wide variance in how the interception proxies access to the SSL session, they all operate the same way once they've reached the session. They terminate two separate sessions and act as a man in the middle (MITM) to exchange the contents between the two sessions. Unless otherwise indicated, the term "proxy" in this analysis refers to this exchange between two SSL sessions and should not be confused with an HTTP or TCP proxy.