What did Hlido score Dnaerys 1000 Genomes MCP?

Dnaerys 1000 Genomes MCP scored 63/100 (FADING) in Hlido's independent, hands-on review.

Does any vendor pay Hlido for placement?

No. Hlido takes no money from the agents it rates — scoring weights stay private and the evidence behind every verdict is public.

Dnaerys 1000 Genomes MCP

Name: Dnaerys 1000 Genomes MCP review
Item: Dnaerys 1000 Genomes MCP
Rating: 63
Author: Hlido Editor

Specialized verticals · tested 2026-08-10 · re-test due 2026-11-10 · by the Hlido desk, not the vendor

In short: Natural-language MCP access to the 1000 Genomes dataset, fronting a specialised in-memory variant store — serious engineering argument, but the captured surface documents the database, not the agent.

4 PASS · 2 FAIL of 6 public-surface claims

Visit Dnaerys 1000 Genomes MCP → scorecard.json

Quick answer

Dnaerys 1000 Genomes MCP scores 63/100 (FADING) on Hlido’s independent, hands-on test (reviewed 2026-08-10). FADING (63) because the underlying variant store publishes a genuine, specific technical argument with documentation depth, but the MCP layer that is actually under review is undocumented on the captured surface — no too

Hlido Editor · 2026-08-10

The reviewed artefact is an MCP server offering natural-language queries over the 1000 Genomes Project dataset (3,202 samples, GRCh38, sequenced by the New York Genome Center), hosted on the Dnaerys variant store. The public surface we could read is the variant store's own documentation, and on its own terms it is unusually substantive: a clearly stated thesis that general-purpose DBMS architectures are the wrong shape for low-latency variant queries, argued through query-plan optimisation, storage-model efficiency and data-allocation latency, with navigation for performance, RAM and disk footprints, operational architecture and licensing. That is a technical argument made in public with enough specificity to be disagreed with, which is more than most infrastructure sites offer. The gap is the review's subject. Nothing on this surface characterises the MCP layer — what tools it exposes, how natural language maps onto variant queries, what it does when a query is ambiguous, or what guardrails exist around genomic data access. A reviewer can see that the substrate is thoughtfully built and cannot see whether the agent interface on top of it is. Public genomic reference data lowers the privacy stakes considerably, which is the reason this is not scored lower.

Why FADING

FADING (63) because the underlying variant store publishes a genuine, specific technical argument with documentation depth, but the MCP layer that is actually under review is undocumented on the captured surface — no tool list, no query semantics, no handling of ambiguous natural-language input. Scoring an agent interface requires seeing the agent interface. The dataset is public reference data rather than patient data, and the documentation is open and login-free, which keeps this well above FLATLINE.

Public-surface checklist

PASS Homepage loads (required)
PASS Primary value prop (required) — 'Low Latency Genome Variant Store'; MCP repo offers NL access to 1000 Genomes
FAIL Cta present (required) — Documentation navigation only; no primary action on the captured page
PASS Pricing or access (required) — 'Licensing & Support' and 'Online Services' in nav
FAIL Evidence or demo (required) — Architectural argument published; no benchmark results or MCP tool documentation captured
PASS Docs public — Full documentation tree public, no login

What we saw

4 screenshots captured by the Hlido engine during the reviewed run (run-8e8452f486ec3f98-dnaerys-org). Our own captures — not vendor marketing material.

Dnaerys 1000 Genomes MCP — run screenshot 1 (home.png) — home.png

Dnaerys 1000 Genomes MCP — run screenshot 2 (page__what-is-dnaerys-variant-store.png) — page__what-is-dnaerys-variant-store.png

Dnaerys 1000 Genomes MCP — run screenshot 3 (page_.png) — page_.png

Dnaerys 1000 Genomes MCP — run screenshot 4 (page_.png) — page_.png

What it does well

States a specific, falsifiable engineering thesis rather than generic performance claims
Documentation covers performance, RAM/disk footprints and operational architecture openly
Built on public reference data (1000 Genomes, GRCh38, NYGC-sequenced) with clear provenance
Licensing, terms and support are surfaced in the primary navigation
Purpose-built architecture argument is technically legible and open to disagreement

What it fails at

The MCP layer under review is not documented on the captured surface — no tool list or query semantics
No evidence of how natural language is mapped to variant queries, or what happens on ambiguity
Performance claims are argued architecturally, not demonstrated with published benchmarks
No guidance on correct interpretation of results for non-genomicist agents or users
Scope is one dataset; no statement about extending to other cohorts

Best for

Bioinformatics teams wanting low-latency programmatic access to 1000 Genomes
Agent developers prototyping genomic question-answering against public reference data
Researchers evaluating in-memory variant stores against warehouse-based approaches
Anyone who needs the query path documented at an architectural level before adopting

Not recommended for

Clinical or diagnostic use — this is public research reference data, not a clinical system
Teams needing published benchmarks rather than an architectural argument
Users expecting documented MCP tool semantics before integration
Cohorts other than 1000 Genomes

Pricing & access

Pricing findable on the public surfacePASS 'Licensing & Support' and 'Online Services' in nav (tested 2026-08-10)

Derived from Hlido-held evidence only (engine checklist + editorial text); quotes are verbatim from the scorecard; not vendor-supplied; re-derived daily. Verify current prices on the vendor's pricing page. Last verified 2026-08-10.

Related agents

Agent relevance

API MCP

Agentic-Commerce Readiness 55/100 · INTEGRABLE

Independent readiness for agent delegation & transaction. How it’s scored · check live

An MCP server fronting a hosted variant store, so an agent can query genomic variation in natural language without handling VCFs. The integration path exists and an API is documented for the store itself; the MCP tool surface is not described publicly, so an integrator cannot determine tool names, arguments or failure behaviour before connecting.

Agent-friendly score: 6/10

Evidence

Provides natural-language access to the 1000 Genomes Project dataset via MCP — source (2026-08-10) verified
Dataset is 3,202 samples, GRCh38, sequenced and aligned by New York Genome Center — source (2026-08-10) verified
Dnaerys is a distributed, horizontally scalable, in-memory variant store — source (2026-08-10) verified
Documentation covers performance, footprints, operational architecture and licensing — source (2026-08-10) verified

scorecard.json · transparency passport · registry · methodology

More: compare agents · best of · developer tools · incident registry

Verdict by Hlido Editor, our automated editorial system · Method: public-surface-tier-1+editorial-narrative-v2 · Methodology version 2026.05 · Next review due 2026-11-10

How this page was produced. The scores, claim verdicts and evidence come from automated hands-on testing of the product’s public surface. The written analysis is drafted by an AI system, and pages publish without a person reviewing each one. Hlido publishes this record and answers for it — tell us if anything here is wrong and we will correct it.

Embed this trust badge

Live, always-current independent score — free to embed on your site or README. No vendor pays for placement.

Markdown

[![Hlido trust score](https://hlido.eu/badge/dnaerys-onekgpd-mcp.svg)](https://hlido.eu/check/?agent=dnaerys-onekgpd-mcp)

HTML

<a href="https://hlido.eu/check/?agent=dnaerys-onekgpd-mcp"><img src="https://hlido.eu/badge/dnaerys-onekgpd-mcp.svg" alt="Hlido trust score"></a>