# HLA-Verify (remote · api.hlaverify.com)

HLA nomenclature and match checks against a pinned IPD-IMGT/HLA release. No patient identifiers.

- Trust score: 64/100 (medium)
- Change this week: +14
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-09-29

## Components

- remote · `api.hlaverify.com`: 64/100 (this document), [markdown](https://verifymcp.io/servers/com-hlaverify-hla-verify/api.md), [page](https://verifymcp.io/servers/com-hlaverify-hla-verify/api)

## Channel facts

- Endpoint: `https://api.hlaverify.com/mcp`
- Transports: `streamable-http`
- Auth: `none`
- Version: `1.0.1`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-09-29.

- **Endpoint Security**: 46/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - Authorisation not fully verified: no authorisation is required to call this server, and 2 tool(s) never declared a destructiveHint. The MCP spec treats an absent hint as destructive by default, so we cannot call this surface safe.
  - HTTPS enforcement could not be verified: the plaintext port answered with HTTP 405, which proves neither a plaintext path nor enforcement.
  - HSTS check failed: the Strict-Transport-Security header is absent.
  - DNSSEC check failed: this domain isn't protected by DNSSEC.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 67/100
  - AI-judged instruction clarity (excellent).
  - Context-footprint check failed: tool/resource definitions use about 1910 tokens (~191/item across 10 items; 10 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 43/100
  - Stability observed for 13 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 98/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 95% of tool parameters carry a description.
  - Structured output schemas are declared (100% of tools); any adoption earns full credit.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - We read all 10 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 11 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a current MCP spec version (2026-07-28).

## Install

### How do I install the HLA-Verify MCP server?

HLA-Verify is a hosted endpoint at https://api.hlaverify.com/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add --transport http com-hlaverify-hla-verify 'https://api.hlaverify.com/mcp'
```

### Cursor

```json
{
  "mcpServers": {
    "com-hlaverify-hla-verify": {
      "url": "https://api.hlaverify.com/mcp"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "com-hlaverify-hla-verify": {
      "type": "http",
      "url": "https://api.hlaverify.com/mcp"
    }
  }
}
```

### Codex

```toml
[mcp_servers.com-hlaverify-hla-verify]
url = "https://api.hlaverify.com/mcp"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "com-hlaverify-hla-verify": {
      "type": "remote",
      "url": "https://api.hlaverify.com/mcp",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add com-hlaverify-hla-verify --url 'https://api.hlaverify.com/mcp' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  com-hlaverify-hla-verify:
    url: "https://api.hlaverify.com/mcp"
```

### Netclaw

```json
{
  "McpServers": {
    "com-hlaverify-hla-verify": {
      "Transport": "http",
      "Url": "https://api.hlaverify.com/mcp"
    }
  }
}
```

### Vellum

```bash
assistant mcp add com-hlaverify-hla-verify -t streamable-http -u 'https://api.hlaverify.com/mcp'
```

### Other

```json
{
  "mcpServers": {
    "com-hlaverify-hla-verify": {
      "type": "http",
      "url": "https://api.hlaverify.com/mcp"
    }
  }
}
```

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-09-28 (score 64, +1)

- [functional] We updated how we score, so this day's move reflects our rubric, not a change to the server

### 2026-09-25 (score 63, +1)

- [functional] We updated how we score, so this day's move reflects our rubric, not a change to the server

### 2026-09-24 (score 62, +11)

- [security improvement] Judged manipulation: unverified → pass
- [functional improvement] Schema quality: unverified → excellent

### 2026-09-23 (score 51, +1)

No change was recorded against any check on this day. Stability & Change Management went from 20 to 23. That category is still filling its 30-day observation window: 6 days of observed history at the previous scan, 7 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-09-22 (score 50, −11)

- [security regression] Judged manipulation: pass → unverified
- [security] The server rewrote its instructions, which are the text every model session reads
- [functional regression] Schema quality: excellent → unverified

### 2026-09-21 (score 61, +1)

No change was recorded against any check on this day. Stability & Change Management went from 13 to 17. That category is still filling its 30-day observation window: 4 days of observed history at the previous scan, 5 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-09-19 (score 60, +1)

No change was recorded against any check on this day. Stability & Change Management went from 7 to 10. That category is still filling its 30-day observation window: 2 days of observed history at the previous scan, 3 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-09-18 (score 59, 0)

- [security] The server rewrote its instructions, which are the text every model session reads
- [functional regression] Schema quality: 1567 → 1863
- [functional] New tool “research_access”

## MCP tools (10)

### `verify_text` (~179 tokens)

Scan HLA typing report text, or model output about HLA, for allele-shaped tokens and classify each one: valid / legacy (with modern form) / deleted (with successor) / fabricated. Nomenclature checking against a pinned IPD-IMGT/HLA release, not interpretation of a case. Use on any AI-generated or transcribed content mentioning HLA. Send the HLA content only, with patient identifiers removed first.

Input parameters:

- `text` (string, required): HLA typing report text, or model output about HLA typing, to scan for allele names. Send the HLA content only: strip patient names, medical record numbers, dates of birth, accession and case identifi…

Output parameters:

- `attribution` (string): Data attribution (IPD-IMGT/HLA, CC-BY-ND).
- `clean` (boolean): The guardrail: true only when no token is hallucinated, fabricated_group or deleted. Gate on this before presenting the text.
- `counts` (object): Number of distinct tokens per status.
- `release` (string): IPD-IMGT/HLA release every verdict was computed against.
- `tokens` (array): Each distinct allele-shaped token found, sorted by token.

### `normalize_allele` (~77 tokens)

Normalize one reported HLA allele name (any era) to current 2-field form, with G group, P group, serologic equivalent, and flags.

Input parameters:

- `name` (string, required): One reported HLA allele name, any nomenclature era. An allele string only, never a patient name, medical record number or other identifier.

Output parameters:

- `allele_2field` (string): Current 2-field form, or UNRESOLVABLE. Never present an UNRESOLVABLE name as an allele.
- `current_name` (string): Current full name in the pinned release, or UNRESOLVABLE.
- `flags` (array): e.g. deprecated_name, nonexistent_allele, null_allele.
- `g_group` (string): G group; NONE (no group), AMBIGUOUS (members differ) or UNRESOLVABLE.
- `reported` (string): The input, verbatim.

### `allele_info` (~108 tokens)

Look up one exact name in the pinned release and return what it is: assigned (G/P group, first release, confirmed status, WMDA serology, null flag), valid_prefix (member count and sample), or deleted (successor). Not found if the name has never existed in any release.

Input parameters:

- `name` (string, required): Exact HLA allele name, a lower-resolution prefix, or a deleted name. An allele string only, never a patient name, medical record number or other identifier.

Output parameters:

- `attribution` (string): Data attribution (IPD-IMGT/HLA, CC-BY-ND).
- `confirmed` (boolean): assigned: confirmed (vs unconfirmed) allele.
- `detail` (string): Present only when the name is not assigned in this release (and then no other field is).
- `first_release` (string|null): assigned: first release the exact name appeared in.
- `g_group` (string|null): assigned: G group, or null.
- `ligands` (object): Class I (A/B/C) ligand facts, aggregated over member alleles: 'ambiguous' when members disagree, 'unknown' when no residue data.
- `members_count` (integer): valid_prefix: number of assigned alleles under the prefix.
- `members_sample` (array): valid_prefix: up to 10 member alleles.
- `name` (string)
- `null_allele` (boolean): assigned: true for an N (null, not expressed) allele.
- `p_group` (string|null): assigned: P group, or null.
- `release` (string): IPD-IMGT/HLA release every verdict was computed against.
- `serology` (object): assigned: WMDA serologic equivalents by column (non-empty columns only).
- `status` (string): assigned: an exact allele in this release; valid_prefix: a lower-resolution prefix of assigned alleles; deleted: withdrawn or renamed (see successor).
- `successor` (string|null): deleted: the current name, or null if none.

### `match_score` (~229 tokens)

Count a donor-recipient HLA match by the published counting rules (R1-R6): allele arithmetic over chromosomes, not a donor recommendation. recipient/donor: {"A": ["A*01:01","A*02:01"], "B": [...], ...} (two reported alleles per locus, any nomenclature era; allele strings only, no patient identifiers). framework: 6/6, 8/8, 10/10, 12/12, or antigen. Returns count, per-locus verdicts, GvH/HvG mismatch counts, and flags; unresolvable typing yields 'potential', never a confident count.

Input parameters:

- `donor` (object, required): locus -> up to 4 reported allele names. Allele strings only: never patient names, medical record numbers, dates of birth, or accession or case identifiers.
- `framework` (string)
- `recipient` (object, required): locus -> up to 4 reported allele names. Allele strings only: never patient names, medical record numbers, dates of birth, or accession or case identifiers.

Output parameters:

- `attribution` (string): Data attribution (IPD-IMGT/HLA, CC-BY-ND).
- `count` (string): 'matched/total' over the resolvable loci only, or UNRESOLVABLE when none resolves. Check verdicts for 'potential' loci before quoting it as a confident count.
- `flags` (array): e.g. resolution_insufficient, null_allele, null_allele_mismatch.
- `framework` (string)
- `gvh_mismatches` (integer): Graft-versus-host mismatches over non-potential loci.
- `hvg_mismatches` (integer): Host-versus-graft mismatches over non-potential loci.
- `release` (string): IPD-IMGT/HLA release every verdict was computed against.
- `verdicts` (object): Framework locus -> verdict.

### `check_typing` (~199 tokens)

QC-check one HLA typing (all loci) against the pinned release: resolves every reported allele, flags unresolvable/outdated/locus-mismatched/null alleles, flags too-many/single/homozygous per locus, computes the B-leader (-21 M/T) and KIR-ligand (C1/C2/Bw4) profile, and DRB3/4/5 expected-vs-reported. Nomenclature and internal-consistency checking of the report, not clinical interpretation. typing: {"A": ["A*01:01", "A*02:01"], "B": [...], "DRB1": [...], ...} (any nomenclature era; allele strings only, no patient identifiers).

Input parameters:

- `typing` (object, required): locus -> up to 4 reported allele names. Allele strings only: never patient names, medical record numbers, dates of birth, or accession or case identifiers.

Output parameters:

- `attribution` (string): Data attribution (IPD-IMGT/HLA, CC-BY-ND).
- `counts` (object)
- `drb345` (object|null): DRB3/4/5 expected from DRB1 vs reported; null when DRB1 is not typed.
- `issues` (array)
- `loci` (object): Reported locus key -> one row per reported allele, in input order.
- `profile` (object)
- `release` (string): IPD-IMGT/HLA release every verdict was computed against.
- `valid` (boolean): true when there are no error-severity issues. Gate on this before using the typing.

### `donor_compat` (~212 tokens)

Donor/recipient immunogenetic compatibility under two published rule sets: HLA-B leader match (-21 M/T, Petersdorf 2020) for a single HLA-B mismatch, and KIR ligand (C1/C2/Bw4) class comparison, computed over each side's full typing QC. Rule checking against published frameworks; it does not rank or recommend a donor. recipient/donor: {"A": [...], "B": [...], "C": [...], "DRB1": [...], ...} (allele strings only, no patient identifiers). Decision support only; not a medical device.

Input parameters:

- `donor` (object, required): locus -> up to 4 reported allele names. Allele strings only: never patient names, medical record numbers, dates of birth, or accession or case identifiers.
- `recipient` (object, required): locus -> up to 4 reported allele names. Allele strings only: never patient names, medical record numbers, dates of birth, or accession or case identifiers.

Output parameters:

- `attribution` (string): Data attribution (IPD-IMGT/HLA, CC-BY-ND).
- `b_leader` (object)
- `donor_valid` (boolean): Donor typing QC had no errors.
- `issues` (object)
- `kir_ligands` (object)
- `recipient_valid` (boolean): Recipient typing QC had no errors.
- `release` (string): IPD-IMGT/HLA release every verdict was computed against.

### `validate_gl_string` (~127 tokens)

Validate and normalize a GL String (Genotype List, ^ | + ~ / grammar): resolves every allele token, flags outdated/unresolvable names and structural problems (mixed loci within a slash-list, a repeated locus within a haplotype or across ^ blocks, more than two haplotypes, differing loci across a genotype or genotype list, empty elements), and returns the normalized string. Grammar and nomenclature checking only; send allele names, not patient identifiers.

Input parameters:

- `gl` (string, required): GL String to validate and normalize. Allele names and GL grammar only, never patient identifiers.

Output parameters:

- `alleles` (array): Each distinct allele token, in first-seen order.
- `attribution` (string): Data attribution (IPD-IMGT/HLA, CC-BY-ND).
- `changed` (boolean): normalized_gl differs from the trimmed input.
- `counts` (object)
- `issues` (array)
- `loci` (array)
- `normalized_gl` (string): The GL String with outdated names replaced by current ones.
- `release` (string): IPD-IMGT/HLA release every verdict was computed against.
- `valid` (boolean): true when there are no error-severity issues.

### `beta_signup` (~140 tokens)

Put a user on the free public beta's notification list for paid API keys. Ask before calling: it records the address they give you. Re-signing the same address is safe (status already_recorded). Someone who needs a higher rate limit today should email hello@hlaverify.com for a beta key instead of waiting.

Input parameters:

- `email` (string, required): The user's email address.
- `org` (string): Lab, company or institution (optional).
- `source` (string): Where the signup came from, e.g. mcp (optional).
- `use_case` (string): What they would use the API for (optional). No patient details.

Output parameters:

- `message` (string): What to tell the user, including how to get a beta key today.
- `ok` (boolean): Always true; a rejected signup comes back as an error result.
- `release` (string): IPD-IMGT/HLA release every verdict was computed against.
- `status` (string): already_recorded: the address was already on the list. Both are success — do not retry.

### `research_access` (~268 tokens)

Apply for free HLA-Verify access for an academic or nonprofit lab. Ask before calling: it records the address, institution and use case you give it. Approval is manual: a person reads every application, so it is not instant and not guaranteed. If it is approved the applicant is emailed a single-use code that takes 100% off a subscription for 12 months at self-serve checkout, with no card and no contract. Re-applying with the same address is safe (status already_recorded) and never overwrites an application that has already been decided. Commercial labs should buy a tier at https://api.hlaverify.com/pricing instead.

Input parameters:

- `email` (string, required): The applicant's email address. The approval code is sent here.
- `expected_volume` (string): Rough call or typing volume, e.g. 'about 20,000 typings a month' (optional).
- `institution` (string, required): University, hospital, institute or nonprofit the work is done at.
- `source` (string): Where the application came from, e.g. mcp (optional).
- `use_case` (string, required): What the research or teaching is, and what the API would be used for. This is what the decision is made on, so be specific. No patient details.

Output parameters:

- `message` (string): What to tell the user, including that approval is by hand and not instant.
- `ok` (boolean): Always true; a rejected application comes back as an error result.
- `release` (string): IPD-IMGT/HLA release every verdict was computed against.
- `status` (string): already_recorded: this address already has an application on file. Both are success; do not retry.

### `about` (~35 tokens)

What this server is and is not, what to send it, benchmark evidence for why to use it, the beta state, and terms.

Output parameters:

- `agents` (string)
- `api` (string)
- `beta` (string)
- `beta_key` (string)
- `beta_signup` (string)
- `code` (string)
- `commercial` (string)
- `demo` (string)
- `disclaimer` (string)
- `inputs` (string)
- `limits` (string)
- `name` (string)
- `release` (string): IPD-IMGT/HLA release every verdict was computed against.
- `research` (string)
- `scope` (string)
- `why` (string)

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/com-hlaverify-hla-verify/api#diagnostics

## Score history

- 2026-09-29: 64
- 2026-09-28: 64
- 2026-09-27: 63
- 2026-09-26: 63
- 2026-09-25: 63
- 2026-09-24: 62
- 2026-09-23: 51
- 2026-09-22: 50
- 2026-09-21: 61
- 2026-09-20: 60
- 2026-09-19: 60
- 2026-09-18: 59
- 2026-09-17: 59
- 2026-09-16: 58

## Common questions

### What is the HLA-Verify MCP server?

HLA-Verify is an MCP server listed in the public MCP registry as com.hlaverify/hla-verify. HLA nomenclature and match checks against a pinned IPD-IMGT/HLA release. No patient identifiers. This page covers its hosted endpoint (https://api.hlaverify.com/mcp).

### Is the HLA-Verify MCP server safe to use?

HLA-Verify scores 64 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the HLA-Verify MCP server expose?

HLA-Verify exposes 10 tools: verify_text, normalize_allele, allele_info, match_score, check_typing, and 5 more. Their descriptions and schemas cost roughly 1,574 tokens of context every time the server is loaded.

### Does the HLA-Verify MCP server require authentication?

No. We connected to HLA-Verify without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

### Is the HLA-Verify MCP server still maintained?

HLA-Verify is still listed as active in the MCP registry. We last reached this channel on 29 September 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

## Links

- Remote endpoint: https://api.hlaverify.com/mcp
- Repository: https://github.com/jasonbrelsford/verifiable-science-envs
- Website: https://hlaverify.com/
- Changelog RSS feed: https://verifymcp.io/servers/com-hlaverify-hla-verify/api.xml
- Changelog JSON feed: https://verifymcp.io/servers/com-hlaverify-hla-verify/api.json
- HTML version of this page: https://verifymcp.io/servers/com-hlaverify-hla-verify/api