Posted on Sep 30 ## Your agent says "done" — and nothing ran Have you ever shipped something an agent said it had finished, and only found out weeks later that nothing had actually run? I have. The failures that cost me time were never the crashes: a crash is a gift, because CI sees it. The expensive ones exit 0, print nothing unusual, and quietly skipped the work. So instead of trusting exit codes I went looking for them in recorded traffic. Everything below is measured: one small agent task — fetch headlines, write a draft, verify it through a tool — running in three frameworks behind a…