The repo's own test
From its README, section Does it produce games, or documents? Four games were built from one brief at different modes.rigor levels:
| rigor | Docs before code | Time |
|---|
| minimal | 1 | 0 min |
| standard | 30 | 58 min |
All four produced a running, tested game. Two blind reviewers and a human playtest ranked the standard build last.