---
title: "Classify every allele-shaped token in HLA report text or model output"
method: POST
path: "/v1/verify"
---

# Classify every allele-shaped token in HLA report text or model output

`POST /v1/verify`

## Request

- Base URL: `https://api.hlaverify.com`
- URL: `https://api.hlaverify.com/v1/verify`
- Auth: one of:
  - none
  - API key in header `X-API-Key`

Nomenclature checking against the pinned release, not interpretation of a case. Send HLA typing report text or model output about HLA typing with patient identifiers removed first: no names, medical record numbers, dates of birth, accession or case identifiers, or other patient details. The service neither needs nor wants them and does not store request bodies; de-identifying is the caller's responsibility.

## Request body

- object
  - `text` string, required: HLA typing report text, or model output about HLA typing, to scan for allele names. Allele names and HLA content only, never patient identifiers.

## Response `200`

verdicts

- object
  - `release` string
  - `clean` boolean
  - `counts` object
  - `tokens` object[]
    - `token` string
    - `status` 'valid' | 'group' | 'deleted' | 'fabricated_group' | 'hallucinated' | 'mac_code' | 'out_of_scope'
    - `note` string
    - `successor` string
    - `current_2field` string
    - `g_group` string
    - `flags` string[]
    - `resolution` object: valid tokens with fewer than four fields: {fields, level, distinguishes} — what this resolution does and does not distinguish.
  - `attribution` string

## Other responses

- `422`: invalid input

---

[API](https://skmtc.dev/hlaverify/apis/hla-verify.md) · [All operations](https://skmtc.dev/hlaverify/apis/hla-verify/llms.txt) · [OpenAPI document](https://skmtc.dev/hlaverify/apis/hla-verify/revisions/40cd5e7d134d?raw)