← back to paper

Review history

arxiv: 2605.20023 · 2 revisions

When Skills Don't Help: A Negative Result on Procedural Knowledge for Tool-Grounded Agents in Offensive Cybersecurity

1. 2026-06-30 UNVERDICTED LOW v0.9.1-grok novelty 5.0
   22618 ms 5884 in 1391 out 2026-06-30T18:08:24.067499+00:00
2. 2026-05-20 CONDITIONAL LOW v0.9.0 novelty 5.0
   82443 ms 5890 in 1348 out 2026-05-20T05:24:22.571363+00:00