Back to Search-quality baseline: prose / conceptual (Phase 1.5, v1.2.0 candidate)
Search-quality baseline: prose / conceptual (Phase 1.5, v1.2.0 candidate)

Method Recap

15 prose queries, each with a regex matching a per-query enumerated valid-URI set. For each: run cupertino search "<query>" --limit 10, find first-relevant rank, compute P@3, P@5, any-match-in-top-3 (binary). Aggregate as means. No paired test (single-system).

Harness source: /tmp/cupertino-search-eval-prose.py. Full JSON dump: /tmp/cupertino-search-eval-prose-20260520.json.