PentestJudge: Judging Agent Behavior against Operational Requirements

We introduce PentestJudge, a system for evaluating the operations of penetration testing agents. PentestJudge is a large language model LLM-as-judge with access to tools that allow it to consume arbitrary trajectories of agent states and tool call history to determine whether a security agent's...

Mozilla Proposes Change to Handling of Subordinate CA Certificates

Mozilla is considering a change to the way that it handles certificates issued by externally operated sub-CAs in an effort to gain more control of how these CAs issue certificates and what those certificates can do. The proposal would involve some new controls to help verify that certificates are...