Measured · Building
PSEval benchmark
The moat. Models age out in a year; yardsticks do not.
What it is
The scorer and held-out benchmark for PowerShell against the Standard — parse gate, surface validator, analyzer, AST checks, voice and style lint. 259 tests, version-aware, run against live Windows manifests.
Why it's the moat
The most valuable unbuilt artifact is a field survey — community models, frontier models, and PSEng on one table. Whoever runs that owns the yardstick regardless of where their own model places.
Not available yet — follow along on the system map.