Jul 2026
Baselines Before Architecture: Evaluating Coding Agents for Autonomous Penetration Testing
A controlled study on the 104-task XBOW benchmark using default coding CLI agents as plain-agent baselines to identify the strongest same-model baseline and tests whether security-specific prompt variants improve its observed score.
Ananda Dhakal, Krishna Neupane, Aarjan Chaudhary
· arXiv.org · 2 citations
· ⚡1