Introducing public-key infrastructure for siebogjs javascript-based agent middleware

Abstract

© 2018 Copyright held by the owner/author(s). Software agents are special type of software entities which are usually used in the field of distributed artificial intelligence, because of their capability to communicate, and their reactive and pro-active behaviour. An important factor in such an environment is security. In this paper, we address the security aspect of the SiebogJS agent middleware in regard to protecting the agent’s programme code and/or payload. For this purpose we use the Public key infrastructure and its related concepts. These enable digital signing of programme code using digital certificates and thus prevent an attacker to inject or include malicious code into an agent or its payload. Additionally, we discuss challenges that arise when employing PKI in such an agent framework.