Glossary
Policy Enforcement Point (PEP)
A policy enforcement point is the place in a system where a decision becomes real. It does not decide whether an action is allowed; it receives a decision from a policy decision point and either lets the action proceed or blocks it. The distinction matters because it separates two failure modes that are easy to conflate: a system can have excellent policy logic and still be insecure if the enforcement point can be bypassed, skipped, or placed after the action it was meant to gate.
Good enforcement points share a few properties. They sit before the effect, not after it, so a blocked action never partially executes. They cannot be bypassed by calling the protected resource through a different path. And their coverage is explicit and testable: for a given system, you can name every consequential action and point to the enforcement point in front of it, rather than trusting that enforcement happens somewhere upstream.
A common design mistake is to enforce policy only at the edge of a system (an API gateway, for example) while internal components trust each other implicitly. That works until an internal component is compromised or misused, at which point there is no enforcement point left between the attacker and the action.