Live registry snapshot
Coverage overview
Roll-up view
Categories and source coverage
Scores are intentionally empty until real model result ingestion exists; coverage, source counts, confidence, and benchmark counts are already visible.
Benchmark catalog
Benchmarks by category
| Benchmark | Category | Subcategory | Objectivity | Sources | Triage |
|---|
Evidence layer
Sources, authority, and what is pulled
Deterministic ingestion
Webpage pull findings
These records show what the fetcher pulled, what it matched, and whether AI review is needed.