AI assistants are entering transplant and immunogenetics workflows — and they fabricate allele names, resurrect deleted ones, and slip into 1990s formats. HLA-Verify checks every allele-shaped token in a typing report, EHR fragment, or model output against a pinned IPD-IMGT/HLA release. No LLM anywhere in the loop: every verdict is computed, versioned, and reproducible.
A 7B open model fabricates allele names in 9% of benchmark tasks and answers 62% of them wrongly with high confidence. Our benchmark, HLA-Bench-A, measures it; this engine catches it at runtime.
Pinned to an exact IPD-IMGT/HLA release (46,652 alleles in 3.65.0, ~600 added per quarter). Deleted names resolve to their successors across 110 releases of history.
Reports are processed in memory and discarded; we log counts, not content. Reference data are fetched from the official source under CC-BY-ND and never redistributed.
The demo runs entirely client-side — nothing you paste leaves your machine. Keyed API for labs and LIMS vendors: pilot slots open.