mcpbeat

Raras MCP Server

io.github.rarasAI/raras
answering

Raras is answering right now. Last checked 9 min ago. It exposes 21 tools. Last commit 21 May 2026.

Disease Twin — Brazil's rare-disease knowledge graph (10,468) over MCP. Source-grounded, PT-BR.

Uptime history 41 hours of history
41 hours agonow
100.0%
Uptime 24h
91 of 91 checks
21
Tools
read from the server
1041 ms
Response time
average over 24h
0
Stars
last commit 21 May 2026

Connect this server

Endpoint below is the one we actually reach during checks — not the one copied from a README. Last verified 9 min ago.

run in your terminal
claude mcp add raras --transport http https://raras.org/api/mcp
~/Library/Application Support/Claude/claude_desktop_config.json
{
  "mcpServers": {
    "raras": {
      "url": "https://raras.org/api/mcp"
    }
  }
}
~/.codex/config.toml
[mcp_servers.raras]
url = "https://raras.org/api/mcp"
.cursor/mcp.json
{
  "mcpServers": {
    "raras": {
      "url": "https://raras.org/api/mcp"
    }
  }
}
.vscode/mcp.json
{
  "mcpServers": {
    "raras": {
      "url": "https://raras.org/api/mcp"
    }
  }
}

Available tools 21

Read directly from the server with tools/list, grouped by what they act on. If a tool disappears, we record the date.

diseases
find_diseases_by_phenotypes
Match por IDs HPO. Use para diagnóstico diferencial.
search_diseases
Full-text search em 10.468 doenças raras. Suporta filtros SUS/trials.
papers
find_papers_for_disease
Literatura relacionada a uma doença via embeddings (:SEMANTIC_MATCH, fallback vetorial).
search_papers_semantic
Busca vetorial sobre 424k papers do PubMed. Use caso clínico, pergunta ou conjunto de sintomas.
active
find_active_trials
Trials ativos com filtro Brazil.
analyze
analyze_clinical_case
Caso clínico → doenças candidatas + literatura relevante em 1 chamada. Apenas informativo, nunca diagnóstico.
communities
find_communities
Comunidades FEBRARARAS por doença.
disease
get_disease_detail
Fenótipos HPO, genes, SUS, trials, descrição clínica em PT.
evidence
get_evidence
Cross-references de autoridade (Orphanet, MONDO, OMIM, MeSH, GARD, UMLS), status de verificação e PMIDs de uma doença. A base da regra "nunca inventar".
explain
explain_relation
Raciocínio fundamentado sobre o grafo: dada uma pergunta (ex.: "por que o gene X se associa à doença Y?", "como a droga Z se relaciona a esta doença?"), retorna caminhos de evidência multi-hop (doença↔gene↔via↔droga↔fenótipo) com proveniência citável. Se não houver caminho no grafo, recusa — nunca inventa.
graph
get_graph_stats
Contagens e métricas do dataset.
hypotheses
get_hypotheses
Candidatos a reposicionamento de fármacos gerados por associação de genes (guilt-by-association): drogas que tratam outras doenças que compartilham genes com esta. HIPÓTESES para investigação, NUNCA recomendações clínicas — cada uma rastreável (via gene X, da doença Y).
literature
get_literature_relations
Relações tipadas extraídas da literatura biomédica (associate/treat/cause/inhibit/stimulate/prevent) entre a doença e genes/fármacos, com nº de publicações como proveniência. Filtrável por predicado.
phenotypes
search_phenotypes
Lookup HPO em PT ou EN. Retorna IDs HP:.
phenotypically
find_phenotypically_similar
Doenças com perfil de fenótipos HPO semelhante (simGIC pré-computado, explicável: nº de fenótipos em comum). Complementa find_similar_diseases (semântico).
recent
get_recent_updates
Doenças cujo gêmeo digital ganhou evidência nova (autoria, PubTator3, Open Targets, verificação) nos últimos N dias. Faça polling para acompanhar mudanças (substituto stateless de subscriptions).
reference
find_reference_centers
Centros por doença e/ou UF, com CNES.
research
get_research_log
Registro cronológico do que o agente autônomo fez por esta doença: autoria, mineração de literatura, evidência genética, verificação, prospecção de fontes, recomputação de relações. Prova de que o gêmeo digital estuda a doença continuamente.
schema
get_schema
Node labels, relationship types e property keys do grafo. Use antes de cypher_query.
similar
find_similar_diseases
Vector similarity via embeddings SOTA.
sus
get_sus_coverage
CEAF, SIGTAP, PNTN, nível de integração.

Endpoints

URLTransportStateLatencyChecked
https://raras.org/api/mcp streamable-http answering 1094 ms 9 min ago

Raras — questions

Answers built from our own checks of this server.

What can Raras do?
It exposes 21 tools, read directly from the server on our last check. Among them: analyze_clinical_case, explain_relation, find_active_trials, find_communities, find_diseases_by_phenotypes, find_papers_for_disease and 15 more. The full list with descriptions is on this page — we take it from the server itself via tools/list, not from a README. How MCP servers expose tools in the first place →
What is Raras mostly used for?
Its tools cluster around papers and diseases. That is what this server is built to work with — the grouping comes from the actual tool names, not from a category we assigned.
Is Raras working right now?
We send a real MCP handshake every 15 minutes. Over the last 24 hours 91 of 91 checks got a reply (100.0%), average response time 1041 ms. The bar chart above shows every period we have measured.
How do I connect Raras?
Copy the ready config from this page — we generate it for Claude Code, Claude Desktop, Codex, Cursor and VS Code, each with the file path that client actually reads. It is a remote server, so there is nothing to install — the client connects to the address.
Does Raras need an API key?
No. Raras completed a full MCP handshake with us as an anonymous client and listed its tools without asking for anything. All 21 of them are readable on this page. This is what we observed, not what the docs claim.
How fast is Raras?
It answers our handshake in 1041 ms on average, which is faster than 4% of all working MCP servers we measure. That is on the slow side — worth knowing if the tool sits inside an interactive loop. The comparison comes from our own checks across the whole registry, every 15 minutes.
Is Raras open source?
Yes — written in TypeScript and 0 stars on GitHub. The source link is on this page, so you can read exactly what it does with your data before you connect it.