In this paper we define a sequent calculus to formally specify, simulate, debug and verify security protocols. In our sequents we distinguish between the current knowledge of principals and the current global state of the session. Hereby, we can describe the operational semantics of principals and of an intruder in a simple and modular way. Furthermore, using proof theoretic tools like the analysis of permutability of rules, we are able to find efficient proof strategies that we prove complete for special classes of security protocols including Needham-Schroeder. Based on the results of this preliminary analysis, we have implemented a Prolog meta-interpreter which allows for rapid prototyping and for checking safety properties of security protocols, and we have applied it for finding error traces and proving correctness of practical examples.