experiment  pr-42 (exp-cand)
suite       support
created     2026-01-02 03:04:05 UTC
model       openai:gpt-4o
git         fedcba987654
runs        5/9 passed (56%), 1 error

Metrics
metric        mean over cases
─────────────────────────────
pass rate                 56%
score.check                 1
score.exact             0.625
score.tone                  1
cost (USD)              0.013
latency (ms)            664.4
steps                       1

Failures (4 of 9 cases)
case      passed  errors  failed evaluators
───────────────────────────────────────────
case-00      0/1          exact
case-01      0/1          exact
case-02      0/1          exact
case-new     0/1       1

╭─ ✗ Gates FAILED  0 passed, 1 failed, 0 skipped ──────────────────────────────────────────────────╮
│         gate          rule     actual  reason                                                    │
│ ────────────────────────────────────────────────────────────────────────                         │
│ ✗ FAIL  success_rate  min 0.8  0.5556  pass_rate 0.5556 is below min 0.8                         │
│ note: only 9 cases evaluated: at least 20 are recommended for meaningful results (ADR 0005); not │
│ a gate failure                                                                                   │
╰──────────────────────────────────────────────────────────────────────────────────────────────────╯
