Claude Code

The 'Claude Code' dataset/benchmark is used to evaluate the performance of LLM-driven automated penetration testing agents against intelligent defense mechanisms in a realistic, turn-based attacker-defender-judge framework.

Papers using Claude Code (3)

TraceLab: Characterizing Coding Agent Workloads for LLM Serving2026IterInject: Indirect Prompt Injection Against LLM Agents via Feedback-Guided Iterative Optimization2026SkCC: Portable and Secure Skill Compilation for Cross-Framework LLM Agents2026