Agent Arena

Operations

Installing, running, controlling cost, storage, troubleshooting.

Running Agent Arena day to day.

Page Covers
installation.md Install, extras, and verifying it works
running-evaluations.md Preflight, trials, concurrency, timeouts
cost-control.md Not spending more than you meant to
storage-and-data.md Where everything lives, and querying it
troubleshooting.md Real error messages, causes and fixes
performance.md What governs wall-clock, and how to speed a sweep up safely