[{"_1":2,"_348":-5,"_349":-5},"loaderData",{"_3":4,"_22":23,"_264":265},"root",{"_5":6,"_7":8},"toasts",[],"visitor",{"_9":10,"_11":12},"viewportWidth",1280,"visitorSession",{"_13":14,"_16":17,"_18":-5,"_19":20,"_21":-5},"attributes",[15],"bot","id",19346472,"membership","referenceId","0199f420-7955-7299-8090-ea86aef37824","userAccount","routes/_public/mcp/servers/~namespace/~slug/_pages/_layout",{"_24":25,"_78":-5,"_79":38,"_27":80,"_180":181,"_262":263},"availableLocales",[26,38,44,50,57,64,71],{"_27":28,"_31":35,"_36":37},"mcpServer",{"_29":30,"_31":32,"_33":34},"description","Bietet LLMs über das Model Context Protocol strukturierten Zugriff auf wichtige biomedizinische Datenbanken, darunter PubTator3 (PubMed/PMC), ClinicalTrials.gov und MyVariant.info.","name","BioMCP","readme","# BioMCP: Biomedical Model Context Protocol\n\nBioMCP ist ein Open-Source-Toolkit (MIT-Lizenz), das KI-Assistenten und -Agenten mit spezialisiertem biomedizinischem Wissen ausstattet. Es basiert auf dem Model Context Protocol (MCP) und verbindet KI-Systeme mit maßgeblichen biomedizinischen Datenquellen. So können sie Fragen zu klinischen Studien, wissenschaftlicher Literatur und Genomvarianten präzise und umfassend beantworten.\n\n[![▶️ Video ansehen](./docs/blog/images/what_is_biomcp_thumbnail.png)](https://www.youtube.com/watch?v=bKxOWrWUUhM)\n\n## Warum BioMCP?\n\nGroße Sprachmodelle verfügen zwar über ein breites allgemeines Wissen, es fehlt ihnen jedoch oft an spezialisierten, fachspezifischen Informationen oder dem Zugriff auf aktuelle Ressourcen. BioMCP schließt diese Lücke für die Biomedizin durch:\n\n* Bereitstellung **eines strukturierten Zugangs** zu klinischen Studien, biomedizinischer Literatur und genomischen Varianten\n* Ermöglichung von **Abfragen in natürlicher Sprache** an spezialisierte Datenbanken, ohne dass Kenntnisse der spezifischen Syntax erforderlich sind\n* Unterstützung **biomedizinischer Forschungsabläufe** durch eine einheitliche Schnittstelle\n* Funktioniert als **MCP-Server** für KI-Assistenten und -Agenten\n\n## Biomedizinische Datenquellen\n\nBioMCP lässt sich in drei wichtige biomedizinische Datenquellen integrieren:\n\n* **PubTator3/PubMed** – Biomedizinische Literatur mit Entitätsanmerkungen\n* **ClinicalTrials.gov** – Datenbank für klinische Studienregister und Ergebnisse\n* **MyVariant.info** – Konsolidierte genetische Variantenannotation aus mehreren Datenbanken\n\n## Verfügbare MCP-Tools\n\n### PubMed und PubTator3\n\n* `article_searcher` : Suche nach Artikeln nach Genen, Krankheiten, Varianten oder Schlüsselwörtern\n* `article_details` : Erhalten Sie detaillierte Artikelinformationen, einschließlich Abstracts und Volltext\n\n### ClinicalTrials.gov\n\n* `trial_searcher` : Erweiterte Studiensuche mit Filterung nach Zustand, Intervention, Phase usw.\n* `trial_protocol` : Detaillierte Informationen zum Testprotokoll\n* `trial_locations` : Standorte und Kontaktinformationen der Studienzentren\n* `trial_outcomes` : Ergebnisse und Ergebnismaße\n* `trial_references` : Verwandte Veröffentlichungen\n\n### MyVariant.info\n\n* `variant_searcher` : Suche nach genetischen Varianten mit ausgefeilter Filterung\n* `variant_details` : Umfassende Anmerkungen aus mehreren Quellen (CIViC, ClinVar, COSMIC, dbSNP usw.)\n\n## Schnellstart\n\n### Für Claude Desktop-Benutzer\n\n1. **Installieren Sie `uv`** , falls Sie es nicht haben (empfohlen):\n\n ```bash\n # MacOS\n brew install uv\n\n # Windows/Linux\n pip install uv\n ```\n\n2. **Konfigurieren Sie Claude Desktop** :\n\n * Öffnen Sie die Claude Desktop-Einstellungen\n * Navigieren Sie zum Abschnitt „Entwickler“\n * Klicken Sie auf „Konfiguration bearbeiten“ und fügen Sie hinzu: GXP2\n * Starten Sie Claude Desktop neu und beginnen Sie mit dem Chatten über biomedizinische Themen!\n\n### Installation des Python-Pakets\n\n```bash\n# Using pip\npip install biomcp-python\n\n# Using uv (recommended for faster installation)\nuv pip install biomcp-python\n\n# Run directly without installation\nuv run --with biomcp-python biomcp trial search --condition \"lung cancer\"\n```\n\n## Befehlszeilenschnittstelle\n\nBioMCP bietet eine umfassende CLI für die direkte Datenbankinteraktion:\n\n```bash\n# Get help\nbiomcp --help\n\n# Run the MCP server\nbiomcp run\n\n# Examples\nbiomcp article search --gene BRAF --disease Melanoma\nbiomcp article get 21717063 --full\nbiomcp trial search --condition \"Lung Cancer\" --phase PHASE3\nbiomcp trial get NCT04280705 Protocol\nbiomcp variant search --gene TP53 --significance pathogenic\nbiomcp variant get rs113488022\n```\n\n## Testen und Verifizieren\n\nTesten Sie Ihr BioMCP-Setup mit dem MCP Inspector:\n\n```bash\nnpx @modelcontextprotocol/inspector uv run --with biomcp-python biomcp run\n```\n\nDadurch wird eine Weboberfläche geöffnet, in der Sie alle verfügbaren Tools erkunden und testen können.\n\n## Enterprise-Version: OncoMCP\n\nOncoMCP erweitert BioMCP mit der unternehmensweiten Präzisions-Onkologie-Plattform (POP) von GenomOncology und bietet:\n\n* **HIPAA-konforme Bereitstellung** : Sichere Vor-Ort-Optionen\n* **Echtzeit-Testabgleich** : Aktueller Status und Abgleich auf Armebene\n* **Integration im Gesundheitswesen** : Nahtlose EHR- und Data Warehouse-Konnektivität\n* **Kuratierte Wissensdatenbank** : Über 15.000 Studien und FDA-Zulassungen\n* **Ausgefeiltes Patienten-Matching** : Verwendung integrierter klinischer und molekularer Profile\n* **Erweiterte NLP** : Strukturierte Extraktion aus unstrukturiertem Text\n* **Umfassende Biomarker-Verarbeitung** : Mutations- und Regelverarbeitung\n\nMehr erfahren: [GenomOncology](https://genomoncology.com/)\n\n## MCP-Register\n\n[![Schmiedeabzeichen](https://smithery.ai/badge/@genomoncology/biomcp)](https://smithery.ai/server/@genomoncology/biomcp)\n\n## Dokumentation\n\nEine umfassende Dokumentation finden Sie unter \n\n## BioMCP-Beispiele-Repo\n\nMöchten Sie BioMCP in Aktion sehen?\n\nSchauen Sie sich das Begleitrepository an: 👉 **[biomcp-examples](https://github.com/genomoncology/biomcp-examples)**\n\nEs enthält reale Eingabeaufforderungen, KI-generierte Forschungsberichte und Evaluierungsläufe für verschiedene Modelle. Nutzen Sie es, um Funktionen zu erkunden, Ergebnisse zu vergleichen oder Ihr eigenes Setup zu benchmarken.\n\nHast du selbst ein cooles Beispiel? **Wir freuen uns über deinen Beitrag!** Forke einfach das Repo und sende einen PR mit deinem Experiment.\n\n## Lizenz\n\nDieses Projekt ist unter der MIT-Lizenz lizenziert.","German","tag","de-DE",{"_27":39,"_31":42,"_36":43},{"_29":40,"_31":32,"_33":41},"Provides LLMs with structured access to critical biomedical databases including PubTator3 (PubMed/PMC), ClinicalTrials.gov, and MyVariant.info through the Model Context Protocol.","# BioMCP: Biomedical Model Context Protocol\n\nBioMCP is an open source (MIT License) toolkit that empowers AI assistants and\nagents with specialized biomedical knowledge. Built following the Model Context\nProtocol (MCP), it connects AI systems to authoritative biomedical data\nsources, enabling them to answer questions about clinical trials, scientific\nliterature, and genomic variants with precision and depth.\n\n[![▶️ Watch the video](./docs/blog/images/what_is_biomcp_thumbnail.png)](https://www.youtube.com/watch?v=bKxOWrWUUhM)\n\n## Why BioMCP?\n\nWhile Large Language Models have broad general knowledge, they often lack\nspecialized domain-specific information or access to up-to-date resources.\nBioMCP bridges this gap for biomedicine by:\n\n- Providing **structured access** to clinical trials, biomedical literature,\n and genomic variants\n- Enabling **natural language queries** to specialized databases without\n requiring knowledge of their specific syntax\n- Supporting **biomedical research** workflows through a consistent interface\n- Functioning as an **MCP server** for AI assistants and agents\n\n## Biomedical Data Sources\n\nBioMCP integrates with three key biomedical data sources:\n\n- **PubTator3/PubMed** - Biomedical literature with entity annotations\n- **ClinicalTrials.gov** - Clinical trial registry and results database\n- **MyVariant.info** - Consolidated genetic variant annotation from multiple\n databases\n\n## Available MCP Tools\n\n### PubMed & PubTator3\n\n- `article_searcher`: Search for articles by genes, diseases, variants, or\n keywords\n- `article_details`: Get detailed article information including abstracts and\n full text\n\n### ClinicalTrials.gov\n\n- `trial_searcher`: Advanced trial search with filtering by condition,\n intervention, phase, etc.\n- `trial_protocol`: Detailed trial protocol information\n- `trial_locations`: Trial site locations and contact information\n- `trial_outcomes`: Results and outcome measures\n- `trial_references`: Related publications\n\n### MyVariant.info\n\n- `variant_searcher`: Search for genetic variants with sophisticated filtering\n- `variant_details`: Comprehensive annotations from multiple sources (CIViC,\n ClinVar, COSMIC, dbSNP, etc.)\n\n## Quick Start\n\n### For Claude Desktop Users\n\n1. **Install `uv`** if you don't have it (recommended):\n\n ```bash\n # MacOS\n brew install uv\n\n # Windows/Linux\n pip install uv\n ```\n\n2. **Configure Claude Desktop**:\n - Open Claude Desktop settings\n - Navigate to Developer section\n - Click \"Edit Config\" and add:\n ```json\n {\n \"mcpServers\": {\n \"biomcp\": {\n \"command\": \"uv\",\n \"args\": [\"run\", \"--with\", \"biomcp-python\", \"biomcp\", \"run\"]\n }\n }\n }\n ```\n - Restart Claude Desktop and start chatting about biomedical topics!\n\n### Python Package Installation\n\n```bash\n# Using pip\npip install biomcp-python\n\n# Using uv (recommended for faster installation)\nuv pip install biomcp-python\n\n# Run directly without installation\nuv run --with biomcp-python biomcp trial search --condition \"lung cancer\"\n```\n\n## Command Line Interface\n\nBioMCP provides a comprehensive CLI for direct database interaction:\n\n```bash\n# Get help\nbiomcp --help\n\n# Run the MCP server\nbiomcp run\n\n# Examples\nbiomcp article search --gene BRAF --disease Melanoma\nbiomcp article get 21717063 --full\nbiomcp trial search --condition \"Lung Cancer\" --phase PHASE3\nbiomcp trial get NCT04280705 Protocol\nbiomcp variant search --gene TP53 --significance pathogenic\nbiomcp variant get rs113488022\n```\n\n## Testing & Verification\n\nTest your BioMCP setup with the MCP Inspector:\n\n```bash\nnpx @modelcontextprotocol/inspector uv run --with biomcp-python biomcp run\n```\n\nThis opens a web interface where you can explore and test all available tools.\n\n## Enterprise Version: OncoMCP\n\nOncoMCP extends BioMCP with GenomOncology's enterprise-grade precision oncology\nplatform (POP), providing:\n\n- **HIPAA-Compliant Deployment**: Secure on-premise options\n- **Real-Time Trial Matching**: Up-to-date status and arm-level matching\n- **Healthcare Integration**: Seamless EHR and data warehouse connectivity\n- **Curated Knowledge Base**: 15,000+ trials and FDA approvals\n- **Sophisticated Patient Matching**: Using integrated clinical and molecular\n profiles\n- **Advanced NLP**: Structured extraction from unstructured text\n- **Comprehensive Biomarker Processing**: Mutation and rule processing\n\nLearn more: [GenomOncology](https://genomoncology.com/)\n\n## MCP Registries\n\n[![smithery badge](https://smithery.ai/badge/@genomoncology/biomcp)](https://smithery.ai/server/@genomoncology/biomcp)\n\n\n

\n\n\n## Documentation\n\nFor comprehensive documentation, visit [https://biomcp.org](https://biomcp.org)\n\n## BioMCP Examples Repo\n\nLooking to see BioMCP in action?\n\nCheck out the companion repository:\n👉 **[biomcp-examples](https://github.com/genomoncology/biomcp-examples)**\n\nIt contains real prompts, AI-generated research briefs, and evaluation runs across different models.\nUse it to explore capabilities, compare outputs, or benchmark your own setup.\n\nHave a cool example of your own?\n**We’d love for you to contribute!** Just fork the repo and submit a PR with your experiment.\n\n## License\n\nThis project is licensed under the MIT License.\n","English","en-US",{"_27":45,"_31":48,"_36":49},{"_29":46,"_31":32,"_33":47},"Proporciona a los LLM acceso estructurado a bases de datos biomédicas críticas, incluidas PubTator3 (PubMed/PMC), ClinicalTrials.gov y MyVariant.info a través del Protocolo de Contexto Modelo.","# BioMCP: Protocolo de contexto del modelo biomédico\n\nBioMCP es un conjunto de herramientas de código abierto (licencia MIT) que dota a los asistentes y agentes de IA de conocimiento biomédico especializado. Desarrollado según el Protocolo de Contexto de Modelo (MCP), conecta los sistemas de IA con fuentes de datos biomédicos fiables, lo que les permite responder preguntas sobre ensayos clínicos, literatura científica y variantes genómicas con precisión y profundidad.\n\n[![▶️ Ver el vídeo](./docs/blog/images/what_is_biomcp_thumbnail.png)](https://www.youtube.com/watch?v=bKxOWrWUUhM)\n\n## ¿Por qué BioMCP?\n\nSi bien los Modelos de Lenguaje Grande poseen un amplio conocimiento general, a menudo carecen de información especializada específica del dominio o de acceso a recursos actualizados. BioMCP cubre esta brecha en la biomedicina mediante:\n\n* Proporcionar **acceso estructurado** a ensayos clínicos, literatura biomédica y variantes genómicas.\n* Permitir **consultas en lenguaje natural** a bases de datos especializadas sin necesidad de conocer su sintaxis específica\n* Apoyar los flujos de trabajo **de investigación biomédica** a través de una interfaz consistente\n* Funciona como un **servidor MCP** para asistentes y agentes de IA\n\n## Fuentes de datos biomédicos\n\nBioMCP se integra con tres fuentes de datos biomédicos clave:\n\n* **PubTator3/PubMed** : Literatura biomédica con anotaciones de entidades\n* **ClinicalTrials.gov** - Registro de ensayos clínicos y base de datos de resultados\n* **MyVariant.info** - Anotación consolidada de variantes genéticas de múltiples bases de datos\n\n## Herramientas MCP disponibles\n\n### PubMed y PubTator3\n\n* `article_searcher` : busca artículos por genes, enfermedades, variantes o palabras clave\n* `article_details` : Obtenga información detallada del artículo, incluidos resúmenes y textos completos\n\n### ClinicalTrials.gov\n\n* `trial_searcher` : Búsqueda avanzada de ensayos con filtrado por condición, intervención, fase, etc.\n* `trial_protocol` : Información detallada del protocolo del ensayo\n* `trial_locations` : Ubicaciones de los sitios de prueba e información de contacto\n* `trial_outcomes` : Resultados y medidas de resultados\n* `trial_references` : Publicaciones relacionadas\n\n### MiVariante.info\n\n* `variant_searcher` : busca variantes genéticas con filtrado sofisticado\n* `variant_details` : Anotaciones completas de múltiples fuentes (CIViC, ClinVar, COSMIC, dbSNP, etc.)\n\n## Inicio rápido\n\n### Para usuarios de Claude Desktop\n\n1. **Instalar `uv`** si no lo tienes (recomendado):\n\n ```bash\n # MacOS\n brew install uv\n\n # Windows/Linux\n pip install uv\n ```\n\n2. **Configurar Claude Desktop** :\n\n * Abrir la configuración de Claude Desktop\n * Vaya a la sección Desarrollador\n * Haga clic en \"Editar configuración\" y agregue: GXP2\n * ¡Reinicia Claude Desktop y comienza a chatear sobre temas biomédicos!\n\n### Instalación de paquetes de Python\n\n```bash\n# Using pip\npip install biomcp-python\n\n# Using uv (recommended for faster installation)\nuv pip install biomcp-python\n\n# Run directly without installation\nuv run --with biomcp-python biomcp trial search --condition \"lung cancer\"\n```\n\n## Interfaz de línea de comandos\n\nBioMCP proporciona una CLI integral para la interacción directa con la base de datos:\n\n```bash\n# Get help\nbiomcp --help\n\n# Run the MCP server\nbiomcp run\n\n# Examples\nbiomcp article search --gene BRAF --disease Melanoma\nbiomcp article get 21717063 --full\nbiomcp trial search --condition \"Lung Cancer\" --phase PHASE3\nbiomcp trial get NCT04280705 Protocol\nbiomcp variant search --gene TP53 --significance pathogenic\nbiomcp variant get rs113488022\n```\n\n## Pruebas y verificación\n\nPruebe su configuración BioMCP con el Inspector MCP:\n\n```bash\nnpx @modelcontextprotocol/inspector uv run --with biomcp-python biomcp run\n```\n\nEsto abre una interfaz web donde puedes explorar y probar todas las herramientas disponibles.\n\n## Versión empresarial: OncoMCP\n\nOncoMCP amplía BioMCP con la plataforma de oncología de precisión de nivel empresarial (POP) de GenomOncology, que proporciona:\n\n* **Implementación conforme a HIPAA** : opciones seguras en las instalaciones\n* **Coincidencia de pruebas en tiempo real** : estado actualizado y coincidencia a nivel de brazo\n* **Integración de la atención médica** : Conectividad fluida entre EHR y almacén de datos\n* **Base de conocimientos seleccionada** : más de 15 000 ensayos y aprobaciones de la FDA\n* **Emparejamiento sofisticado de pacientes** : uso de perfiles clínicos y moleculares integrados\n* **PNL avanzada** : Extracción estructurada de texto no estructurado\n* **Procesamiento integral de biomarcadores** : procesamiento de mutaciones y reglas\n\nMás información: [GenomOncology](https://genomoncology.com/)\n\n## Registros MCP\n\n[![insignia de herrería](https://smithery.ai/badge/@genomoncology/biomcp)](https://smithery.ai/server/@genomoncology/biomcp)\n\n## Documentación\n\nPara obtener documentación completa, visite \n\n## Informe de ejemplos de BioMCP\n\n¿Quieres ver BioMCP en acción?\n\nConsulta el repositorio complementario: 👉 **[biomcp-examples](https://github.com/genomoncology/biomcp-examples)**\n\nContiene indicaciones reales, informes de investigación generados por IA y evaluaciones con diferentes modelos. Úselo para explorar capacidades, comparar resultados o evaluar su propia configuración.\n\n¿Tienes un ejemplo genial? **¡Nos encantaría que contribuyeras!** Simplemente bifurca el repositorio y envía un PR con tu experimento.\n\n## Licencia\n\nEste proyecto está licenciado bajo la licencia MIT.","Spanish","es-ES",{"_27":51,"_31":55,"_36":56},{"_29":52,"_31":53,"_33":54},"モデルコンテキストプロトコルを通じて、PubTator3 (PubMed/PMC)、ClinicalTrials.gov、MyVariant.info などの重要な生物医学データベースへの構造化されたアクセスを LLM に提供します。","バイオMCP","# BioMCP: 生物医学モデルコンテキストプロトコル\n\nBioMCPは、AIアシスタントやエージェントに専門的な生物医学知識を提供するオープンソース（MITライセンス）ツールキットです。モデルコンテキストプロトコル（MCP）に基づいて構築されており、AIシステムを信頼できる生物医学データソースに接続することで、臨床試験、科学文献、ゲノム変異に関する質問に正確かつ詳細に回答することを可能にします。\n\n[![▶️ ビデオを見る](./docs/blog/images/what_is_biomcp_thumbnail.png)](https://www.youtube.com/watch?v=bKxOWrWUUhM)\n\n## なぜ BioMCP なのか?\n\n大規模言語モデルは幅広い一般知識を有しているものの、専門的なドメイン固有の情報や最新のリソースへのアクセスが不足していることがよくあります。BioMCPは、以下の方法でバイオメディシンにおけるこのギャップを埋めます。\n\n* 臨床試験、生物医学文献、ゲノム変異への**構造化されたアクセス**を提供する\n* 特定の構文を知らなくても、専門データベースへの**自然言語クエリ**を可能にする\n* 一貫したインターフェースを通じて**バイオメディカル研究**ワークフローをサポート\n* AIアシスタントおよびエージェント用の**MCPサーバー**として機能する\n\n## 生物医学データソース\n\nBioMCP は、次の 3 つの主要な生物医学データソースと統合されます。\n\n* **PubTator3/PubMed** - エンティティ注釈付きの生物医学文献\n* **ClinicalTrials.gov** - 臨床試験登録および結果データベース\n* **MyVariant.info** - 複数のデータベースからの統合された遺伝子変異アノテーション\n\n## 利用可能なMCPツール\n\n### PubMedとPubTator3\n\n* `article_searcher` : 遺伝子、疾患、変異、キーワードで記事を検索します\n* `article_details` : 抄録や全文を含む詳細な記事情報を取得します\n\n### 臨床試験.gov\n\n* `trial_searcher` : 条件、介入、フェーズなどでフィルタリングする高度な試験検索。\n* `trial_protocol` : 詳細な試験プロトコル情報\n* `trial_locations`場所の場所と連絡先情報\n* `trial_outcomes` : 結果と成果指標\n* `trial_references` ：関連出版物\n\n### マイバリアントインフォ\n\n* `variant_searcher` : 洗練されたフィルタリングによる遺伝子変異の検索\n* `variant_details` : 複数のソース (CIViC、ClinVar、COSMIC、dbSNP など) からの包括的なアノテーション\n\n## クイックスタート\n\n### Claudeデスクトップユーザー向け\n\n1. **`uv`がインストールされていない場合はインストールします**(推奨):\n\n ```bash\n # MacOS\n brew install uv\n\n # Windows/Linux\n pip install uv\n ```\n\n2. **Claude デスクトップを構成する**:\n\n * Claudeデスクトップの設定を開く\n * 開発者セクションに移動\n * 「設定の編集」をクリックし、GXP2を追加します。\n * Claude Desktop を再起動して、バイオメディカルのトピックについてチャットを始めましょう。\n\n### Python パッケージのインストール\n\n```bash\n# Using pip\npip install biomcp-python\n\n# Using uv (recommended for faster installation)\nuv pip install biomcp-python\n\n# Run directly without installation\nuv run --with biomcp-python biomcp trial search --condition \"lung cancer\"\n```\n\n## コマンドラインインターフェース\n\nBioMCP は、データベースとの直接的な対話のための包括的な CLI を提供します。\n\n```bash\n# Get help\nbiomcp --help\n\n# Run the MCP server\nbiomcp run\n\n# Examples\nbiomcp article search --gene BRAF --disease Melanoma\nbiomcp article get 21717063 --full\nbiomcp trial search --condition \"Lung Cancer\" --phase PHASE3\nbiomcp trial get NCT04280705 Protocol\nbiomcp variant search --gene TP53 --significance pathogenic\nbiomcp variant get rs113488022\n```\n\n## テストと検証\n\nMCP Inspector を使用して BioMCP の設定をテストします。\n\n```bash\nnpx @modelcontextprotocol/inspector uv run --with biomcp-python biomcp run\n```\n\nこれにより、利用可能なすべてのツールを調べてテストできる Web インターフェイスが開きます。\n\n## エンタープライズ版: OncoMCP\n\nOncoMCP は、GenomOncology のエンタープライズグレードの高精度腫瘍学プラットフォーム (POP) を活用して BioMCP を拡張し、以下を提供します。\n\n* **HIPAA準拠の導入**：安全なオンプレミスオプション\n* **リアルタイムトライアルマッチング**：最新のステータスとアームレベルのマッチング\n* **ヘルスケア統合**：シームレスなEHRとデータウェアハウスの接続\n* **厳選された知識ベース**：15,000件以上の試験とFDA承認\n* **高度な患者マッチング**：統合された臨床プロファイルと分子プロファイルの使用\n* **高度なNLP** ：非構造化テキストからの構造化抽出\n* **包括的なバイオマーカー処理**：変異とルール処理\n\n詳細はこちら： [GenomOncology](https://genomoncology.com/)\n\n## MCPレジストリ\n\n[![鍛冶屋のバッジ](https://smithery.ai/badge/@genomoncology/biomcp)](https://smithery.ai/server/@genomoncology/biomcp)\n\n## ドキュメント\n\n包括的なドキュメントについては、をご覧ください。\n\n## BioMCP サンプルリポジトリ\n\nBioMCP の動作を実際にご覧になりたいですか?\n\nコンパニオンリポジトリをチェックしてください: 👉 **[biomcp-examples](https://github.com/genomoncology/biomcp-examples)**\n\n実際のプロンプト、AIが生成した調査概要、そして様々なモデル間の評価実行が含まれています。機能の探索、出力の比較、あるいはご自身の設定のベンチマークなどにご活用ください。\n\nご自身のクールな例をお持ちですか？\\*\\*ぜひご協力ください！\\*\\*リポジトリをフォークして、実験内容をPRとして送信してください。\n\n## ライセンス\n\nこのプロジェクトは MIT ライセンスに基づいてライセンスされています。","Japanese","ja-JP",{"_27":58,"_31":62,"_36":63},{"_29":59,"_31":60,"_33":61},"Model Context Protocol을 통해 PubTator3(PubMed/PMC), ClinicalTrials.gov, MyVariant.info를 포함한 중요한 생물의학 데이터베이스에 대한 체계적인 액세스를 LLM에게 제공합니다.","바이오MCP","# BioMCP: 생물의학 모델 컨텍스트 프로토콜\n\nBioMCP는 AI 어시스턴트와 에이전트에게 전문적인 생물의학 지식을 제공하는 오픈 소스(MIT 라이선스) 툴킷입니다. 모델 컨텍스트 프로토콜(MCP)에 따라 구축된 BioMCP는 AI 시스템을 권위 있는 생물의학 데이터 소스에 연결하여 임상 시험, 과학 문헌, 유전체 변이에 대한 질문에 정확하고 심도 있게 답변할 수 있도록 지원합니다.\n\n[![▶️ 영상 보기](./docs/blog/images/what_is_biomcp_thumbnail.png)](https://www.youtube.com/watch?v=bKxOWrWUUhM)\n\n## 왜 BioMCP인가?\n\n대규모 언어 모델은 광범위한 일반 지식을 제공하지만, 전문화된 도메인별 정보나 최신 자료에 대한 접근성이 부족한 경우가 많습니다. BioMCP는 다음과 같은 방법으로 생물의학 분야에서 이러한 격차를 해소합니다.\n\n* 임상 시험, 생물 의학 문헌 및 게놈 변이에 대한 **체계적인 액세스** 제공\n* 특정 구문에 대한 지식이 필요 없이 전문 데이터베이스에 대한 **자연어 쿼리를** 활성화합니다.\n* 일관된 인터페이스를 통해 **생물의학 연구** 워크플로 지원\n* AI 어시스턴트 및 에이전트를 위한 **MCP 서버** 로 기능\n\n## 생물의학 데이터 소스\n\nBioMCP는 세 가지 주요 생물의학 데이터 소스와 통합됩니다.\n\n* **PubTator3/PubMed** - 엔티티 주석이 포함된 생물의학 문헌\n* **ClinicalTrials.gov** - 임상 시험 등록 및 결과 데이터베이스\n* **MyVariant.info** - 여러 데이터베이스의 통합 유전자 변이 주석\n\n## 사용 가능한 MCP 도구\n\n### PubMed 및 PubTator3\n\n* `article_searcher` : 유전자, 질병, 변이 또는 키��드로 기사를 검색합니다.\n* `article_details` : 초록과 전문을 포함한 자세한 기사 정보를 얻으세요\n\n### ClinicalTrials.gov\n\n* `trial_searcher` : 조건, 개입, 단계 등을 기준으로 필터링하여 고급 임상 시험 검색을 제공합니다.\n* `trial_protocol` : 자세한 시험 프로토콜 정보\n* `trial_locations` : 시험장 위치 및 연락처 정보\n* `trial_outcomes` : 결과 및 성과 측정\n* `trial_references` : 관련 출판물\n\n### 마이바리언트.인포\n\n* `variant_searcher` : 정교한 필터링을 통해 유전적 변이를 검색합니다.\n* `variant_details` : 여러 소스(CIViC, ClinVar, COSMIC, dbSNP 등)의 포괄적인 주석\n\n## 빠른 시작\n\n### Claude 데스크톱 사용자를 위한\n\n1. **`uv` 가 없다면 설치하세요** (권장):\n\n 지엑스피1\n\n2. **Claude Desktop 구성** :\n\n * Claude Desktop 설정 열기\n * 개발자 섹션으로 이동\n * \"구성 편집\"을 클릭하고 GXP2를 추가하세요.\n * Claude Desktop을 재시작하고 생물의학 주제에 대한 대화를 시작하세요!\n\n### Python 패키지 설치\n\n```bash\n# Using pip\npip install biomcp-python\n\n# Using uv (recommended for faster installation)\nuv pip install biomcp-python\n\n# Run directly without installation\nuv run --with biomcp-python biomcp trial search --condition \"lung cancer\"\n```\n\n## 명령줄 인터페이스\n\nBioMCP는 직접적인 데이터베이스 상호작용을 위한 포괄적인 CLI를 제공합니다.\n\n```bash\n# Get help\nbiomcp --help\n\n# Run the MCP server\nbiomcp run\n\n# Examples\nbiomcp article search --gene BRAF --disease Melanoma\nbiomcp article get 21717063 --full\nbiomcp trial search --condition \"Lung Cancer\" --phase PHASE3\nbiomcp trial get NCT04280705 Protocol\nbiomcp variant search --gene TP53 --significance pathogenic\nbiomcp variant get rs113488022\n```\n\n## 테스트 및 검증\n\nMCP Inspector로 BioMCP 설정을 테스트하세요.\n\n```bash\nnpx @modelcontextprotocol/inspector uv run --with biomcp-python biomcp run\n```\n\n이렇게 하면 사용 가능한 모든 도구를 탐색하고 테스트할 수 있는 웹 인터페이스가 열립니다.\n\n## 엔터프라이즈 버전: OncoMCP\n\nOncoMCP는 GenomOncology의 엔터프라이즈급 정밀 종양학 플랫폼(POP)으로 BioMCP를 확장하여 다음을 제공합니다.\n\n* **HIPAA 규정 준수 배포** : 안전한 온프레미스 옵션\n* **실시간 시험 매칭** : 최신 상태 및 Arm 레벨 매칭\n* **의료 통합** : 원활한 EHR 및 데이터웨어하우스 연결\n* **큐레이팅된 지식 기반** : 15,000개 이상의 시험 및 FDA 승인\n* **정교한 환자 매칭** : 통합된 임상 및 분자 프로필 사용\n* **고급 NLP** : 비정형 텍스트에서 구조화된 추출\n* **종합적인 바이오마커 처리** : 돌연변이 및 규칙 처리\n\n자세히 알아보기: [GenomOncology](https://genomoncology.com/)\n\n## MCP 레지스트리\n\n[![대장간 배지](https://smithery.ai/badge/@genomoncology/biomcp)](https://smithery.ai/server/@genomoncology/biomcp)\n\n## 선적 서류 비치\n\n자세한 내용은 에서 확인하세요.\n\n## BioMCP 예제 저장소\n\nBioMCP가 실제로 어떻게 활용되는지 보고 싶으신가요?\n\n동반 저장소를 확인하세요: 👉 **[biomcp-examples](https://github.com/genomoncology/biomcp-examples)**\n\n실제 프롬프트, AI가 생성한 연구 요약, 그리고 다양한 모델에 대한 평가 실행이 포함되어 있습니다. 기능을 탐색하고, 결과를 비교하고, 자체 설정을 벤치마킹하는 데 활용하세요.\n\n멋진 예시가 있으신가요? **참여해 주시면 정말 감사하겠습니다!** 저장소를 포크하고 실험 결과를 PR로 제출해 주세요.\n\n## 특허\n\n이 프로젝트는 MIT 라이선스에 따라 라이선스가 부여되었습니다.","Korean","ko-KR",{"_27":65,"_31":69,"_36":70},{"_29":66,"_31":67,"_33":68},"Предоставляет магистрам права структурированный доступ к важнейшим биомедицинским базам данных, включая PubTator3 (PubMed/PMC), ClinicalTrials.gov и MyVariant.info, через протокол контекста модели.","БиоМЦП","# BioMCP: протокол контекста биомедицинской модели\n\nBioMCP — это набор инструментов с открытым исходным кодом (лицензия MIT), который предоставляет помощникам и агентам ИИ специализированные биомедицинские знания. Созданный в соответствии с Model Context Protocol (MCP), он соединяет системы ИИ с авторитетными источниками биомедицинских данных, позволяя им отвечать на вопросы о клинических испытаниях, научной литературе и геномных вариантах с точностью и глубиной.\n\n[![▶️ Посмотрите видео](./docs/blog/images/what_is_biomcp_thumbnail.png)](https://www.youtube.com/watch?v=bKxOWrWUUhM)\n\n## Почему БиоМЦП?\n\nХотя у Large Language Models есть широкие общие знания, им часто не хватает специализированной информации по конкретной области или доступа к актуальным ресурсам. BioMCP устраняет этот пробел для биомедицины:\n\n* Предоставление **структурированного доступа** к клиническим испытаниям, биомедицинской литературе и геномным вариантам\n* Возможность отправлять **запросы на естественном языке** в специализированные базы данных без необходимости знания их специфического синтаксиса\n* Поддержка рабочих процессов **биомедицинских исследований** с помощью единого интерфейса\n* Функционирование в качестве **MCP-сервера** для помощников и агентов на основе искусственного интеллекта\n\n## Источники биомедицинских данных\n\nBioMCP интегрируется с тремя ключевыми источниками биомедицинских данных:\n\n* **PubTator3/PubMed** — Биомедицинская литература с аннотациями сущностей\n* **ClinicalTrials.gov** — Реестр клинических испытаний и база данных результатов\n* **MyVariant.info** — Консолидированная аннотация генетических вариантов из нескольких баз данных\n\n## Доступные инструменты MCP\n\n### PubMed и PubTator3\n\n* `article_searcher` : Поиск статей по генам, заболеваниям, вариантам или ключевым словам\n* `article_details` : Получите подробную информацию о статье, включая аннотации и полный текст\n\n### ClinicalTrials.gov\n\n* `trial_searcher` : Расширенный поиск испытаний с фильтрацией по состоянию, вмешательству, фазе и т. д.\n* `trial_protocol` : Подробная информация о протоколе испытания\n* `trial_locations` : Местонахождение испытательных площадок и контактная информация\n* `trial_outcomes` : Результаты и показатели результатов\n* `trial_references` : Связанные публикации\n\n### MyVariant.info\n\n* `variant_searcher` : Поиск генетических вариантов с помощью сложной фильтрации\n* `variant_details` : Подробные аннотации из нескольких источников (CIViC, ClinVar, COSMIC, dbSNP и т. д.)\n\n## Быстрый старт\n\n### Для пользователей Claude Desktop\n\n1. **Установите `uv`** если у вас его нет (рекомендуется):\n\n ```bash\n # MacOS\n brew install uv\n\n # Windows/Linux\n pip install uv\n ```\n\n2. **Настройте рабочий стол Claude** :\n\n * Откройте настройки Claude Desktop\n * Перейдите в раздел «Разработчик»\n * Нажмите «Изменить конфигурацию» и добавьте: GXP2\n * Перезапустите Claude Desktop и начните общаться на биомедицинские темы!\n\n### Установка пакета Python\n\n```bash\n# Using pip\npip install biomcp-python\n\n# Using uv (recommended for faster installation)\nuv pip install biomcp-python\n\n# Run directly without installation\nuv run --with biomcp-python biomcp trial search --condition \"lung cancer\"\n```\n\n## Интерфейс командной строки\n\nBioMCP предоставляет комплексный интерфейс командной строки для прямого взаимодействия с базой данных:\n\n```bash\n# Get help\nbiomcp --help\n\n# Run the MCP server\nbiomcp run\n\n# Examples\nbiomcp article search --gene BRAF --disease Melanoma\nbiomcp article get 21717063 --full\nbiomcp trial search --condition \"Lung Cancer\" --phase PHASE3\nbiomcp trial get NCT04280705 Protocol\nbiomcp variant search --gene TP53 --significance pathogenic\nbiomcp variant get rs113488022\n```\n\n## Тестирование и проверка\n\nПроверьте настройку BioMCP с помощью MCP Inspector:\n\n```bash\nnpx @modelcontextprotocol/inspector uv run --with biomcp-python biomcp run\n```\n\nОткроется веб-интерфейс, где вы сможете изучить и протестировать все доступные инструменты.\n\n## Корпоративная версия: OncoMCP\n\nOncoMCP расширяет возможности BioMCP с помощью корпоративной платформы GenomOncology для прецизионной онкологии (POP), которая обеспечивает:\n\n* **Развертывание, соответствующее HIPAA** : безопасные локальные варианты\n* **Сопоставление испытаний в реальном времени** : актуальный статус и сопоставление на уровне рук\n* **Интеграция в здравоохранении** : бесперебойное подключение к электронным медицинским картам и хранилищам данных\n* **Проверенная база знаний** : более 15 000 испытаний и одобрений FDA\n* **Сложный подбор пациентов** : использование интегрированных клинических и молекулярных профилей\n* **Расширенный NLP** : структурированное извлечение из неструктурированного текста\n* **Комплексная обработка биомаркеров** : обработка мутаций и правил\n\nУзнайте больше: [ГеномОнкология](https://genomoncology.com/)\n\n## Реестры MCP\n\n[![значок кузнеца](https://smithery.ai/badge/@genomoncology/biomcp)](https://smithery.ai/server/@genomoncology/biomcp)\n\n## Документация\n\nПодробную документацию можно найти на [сайте https://biomcp.org](https://biomcp.org)\n\n## Примеры BioMCP Репозиторий\n\nХотите увидеть BioMCP в действии?\n\nОзнакомьтесь с сопутствующим репозиторием: 👉 **[biomcp-examples](https://github.com/genomoncology/biomcp-examples)**\n\nОн содержит реальные подсказки, сгенерированные ИИ исследовательские брифинги и оценочные прогоны по разным моделям. Используйте его для изучения возможностей, сравнения результатов или бенчмаркинга собственной настройки.\n\nУ вас есть собственный крутой пример? **Мы будем рады, если вы поучаствуете!** Просто сделайте форк репозитория и отправьте PR с вашим экспериментом.\n\n## Лицензия\n\nДанный проект лицензирован по лицензии MIT.","Russian","ru-RU",{"_27":72,"_31":76,"_36":77},{"_29":73,"_31":74,"_33":75},"通过模型上下文协议为 LLM 提供对关键生物医学数据库的结构化访问，包括 PubTator3（PubMed/PMC）、ClinicalTrials.gov 和 MyVariant.info。","生物MCP","# BioMCP：生物医学模型上下文协议\n\nBioMCP 是一个开源（MIT 许可证）工具包，旨在为 AI 助手和代理提供专业的生物医学知识。它基于模型上下文协议 (MCP) 构建，将 AI 系统连接到权威的生物医学数据源，使其能够精准深入地回答有关临床试验、科学文献和基因组变异的问题。\n\n[![▶️观看视频](./docs/blog/images/what_is_biomcp_thumbnail.png)](https://www.youtube.com/watch?v=bKxOWrWUUhM)\n\n## 为什么选择 BioMCP？\n\n虽然大型语言模型拥有广泛的通用知识，但它们往往缺乏特定领域的专业信息或无法获取最新资源。BioMCP 通过以下方式为生物医学弥补了这一差距：\n\n* 提供临床试验、生物医学文献和基因组变异**的结构化访问**\n* 无需了解特定语法即可对专用数据库进行**自然语言查询**\n* 通过一致的界面支持**生物医学研究**工作流程\n* 充当 AI 助手和代理的**MCP 服务器**\n\n## 生物医学数据源\n\nBioMCP 与三个关键生物医学数据源集成：\n\n* \\*\\*PubTator3/PubMed——\\*\\*带有实体注释的生物医学文献\n* \\*\\*ClinicalTrials.gov——\\*\\*临床试验注册和结果数据库\n* **MyVariant.info** - 来自多个数据库的整合遗传变异注释\n\n## 可用的 MCP 工具\n\n### PubMed 和 PubTator3\n\n* `article_searcher` ：按基因、疾病、变异或关键词搜索文章\n* `article_details` ：获取详细的文章信息，包括摘要和全文\n\n### ClinicalTrials.gov\n\n* `trial_searcher` ：高级试验搜索，可按条件、干预、阶段等进行过滤。\n* `trial_protocol` ：详细的试验协议信息\n* `trial_locations` ：试验地点和联系信息\n* `trial_outcomes` ：结果和成果测量\n* `trial_references` ：相关出版物\n\n### MyVariant.info\n\n* `variant_searcher` ：通过复杂的过滤功能搜索遗传变异\n* `variant_details` ：来自多个来源的综合注释（CIViC、ClinVar、COSMIC、dbSNP 等）\n\n## 快速入门\n\n### 对于 Claude 桌面用户\n\n1. 如果没有，**请安装`uv`** （推荐）：\n\n ```bash\n # MacOS\n brew install uv\n\n # Windows/Linux\n pip install uv\n ```\n\n2. **配置Claude桌面**：\n\n * 打开 Claude 桌面设置\n * 导航至开发者部分\n * 点击“编辑配置”并添加：GXP2\n * 重新启动 Claude Desktop 并开始讨论生物医学话题！\n\n### Python 包安装\n\n```bash\n# Using pip\npip install biomcp-python\n\n# Using uv (recommended for faster installation)\nuv pip install biomcp-python\n\n# Run directly without installation\nuv run --with biomcp-python biomcp trial search --condition \"lung cancer\"\n```\n\n## 命令行界面\n\nBioMCP 提供了用于直接数据库交互的综合 CLI：\n\n```bash\n# Get help\nbiomcp --help\n\n# Run the MCP server\nbiomcp run\n\n# Examples\nbiomcp article search --gene BRAF --disease Melanoma\nbiomcp article get 21717063 --full\nbiomcp trial search --condition \"Lung Cancer\" --phase PHASE3\nbiomcp trial get NCT04280705 Protocol\nbiomcp variant search --gene TP53 --significance pathogenic\nbiomcp variant get rs113488022\n```\n\n## 测试与验证\n\n使用 MCP Inspector 测试您的 BioMCP 设置：\n\n```bash\nnpx @modelcontextprotocol/inspector uv run --with biomcp-python biomcp run\n```\n\n这将打开一个 Web 界面，您可以在其中探索和测试所有可用的工具。\n\n## 企业版：OncoMCP\n\nOncoMCP 通过 GenomOncology 的企业级精准肿瘤学平台 (POP) 扩展了 BioMCP，提供：\n\n* **符合 HIPAA 标准的部署**：安全的内部部署选项\n* **实时试验匹配**：最新状态和臂级匹配\n* **医疗保健整合**：无缝 EHR 和数据仓库连接\n* **精选知识库**：15,000 多项试验和 FDA 批准\n* **精准的患者匹配**：使用综合的临床和分子特征\n* **高级 NLP** ：从非结构化文本中进行结构化提取\n* **全面的生物标志物处理**：突变和规则处理\n\n了解更多： [GenomOncology](https://genomoncology.com/)\n\n## MCP 注册中心\n\n[![铁匠徽章](https://smithery.ai/badge/@genomoncology/biomcp)](https://smithery.ai/server/@genomoncology/biomcp)\n\n## 文档\n\n如需完整文档，请访问\n\n## BioMCP 示例库\n\n想要看看 BioMCP 的实际作用吗？\n\n查看配套存储库：👉 **[biomcp-examples](https://github.com/genomoncology/biomcp-examples)**\n\n它包含真实提示、AI 生成的研究摘要以及跨不同模型的评估运行。使用它来探索功能、比较输出结果或对您自己的设置进行基准测试。\n\n你自己有一个很棒的示例吗？\\*\\*我们期待你的贡献！\\*\\*只需 fork 代码库并提交包含你的实验的 PR 即可。\n\n## 执照\n\n该项目已获得 MIT 许可。","Chinese","zh-CN","githubUser","locale",{"_81":82,"_13":83,"_93":40,"_94":40,"_16":95,"_96":97,"_31":32,"_116":117,"_124":-5,"_125":-5,"_126":127,"_168":169,"_174":-5,"_104":144,"_175":176,"_107":177,"_178":179},"addedAt","2025-04-07T18:04:04.177901Z",[84,85,86,87,88,89,90,91,92],"category:bioinformatics","category:research-and-data","category:databases","environment:network","language:python","security_grade:a","quality_grade:a","license_grade:a","author:claimed","descriptionMarkdown","descriptionPlainText",5116,"integrations",[98,109],{"_99":100,"_29":106,"_107":108},"brand",{"_101":102,"_31":103,"_104":105},"logoSvgPath","M8.23 7.982l.006-1.005C7.846 1.417 5.096 0 5.096 0l.048 2.291C3.73 1.056 2.6 1.444 2.6 1.444l.118 15.307s4.218-1.796 5.428 5.505C10.238 13.535 21.401 24 21.401 24V9S10.52-.18 8.231 7.982zm9.79 9.941l-1.046-5.232-1.904 4.507h-.96l-1.72-4.301-1.046 5.04H9.321l2.093-9.39h.802l2.491 5.543 2.508-5.557h.869l2.075 9.39h-2.138z","PubMed","slug","pubmed","Enables searching and retrieval of biomedical literature from PubMed/PMC including article search and full text access through the PubTator3 API.","uid","dmjhgym294",{"_99":110,"_29":114,"_107":115},{"_101":111,"_31":112,"_104":113},"M14.25.18l.9.2.73.26.59.3.45.32.34.34.25.34.16.33.1.3.04.26.02.2-.01.13V8.5l-.05.63-.13.55-.21.46-.26.38-.3.31-.33.25-.35.19-.35.14-.33.1-.3.07-.26.04-.21.02H8.77l-.69.05-.59.14-.5.22-.41.27-.33.32-.27.35-.2.36-.15.37-.1.35-.07.32-.04.27-.02.21v3.06H3.17l-.21-.03-.28-.07-.32-.12-.35-.18-.36-.26-.36-.36-.35-.46-.32-.59-.28-.73-.21-.88-.14-1.05-.05-1.23.06-1.22.16-1.04.24-.87.32-.71.36-.57.4-.44.42-.33.42-.24.4-.16.36-.1.32-.05.24-.01h.16l.06.01h8.16v-.83H6.18l-.01-2.75-.02-.37.05-.34.11-.31.17-.28.25-.26.31-.23.38-.2.44-.18.51-.15.58-.12.64-.1.71-.06.77-.04.84-.02 1.27.05zm-6.3 1.98l-.23.33-.08.41.08.41.23.34.33.22.41.09.41-.09.33-.22.23-.34.08-.41-.08-.41-.23-.33-.33-.22-.41-.09-.41.09zm13.09 3.95l.28.06.32.12.35.18.36.27.36.35.35.47.32.59.28.73.21.88.14 1.04.05 1.23-.06 1.23-.16 1.04-.24.86-.32.71-.36.57-.4.45-.42.33-.42.24-.4.16-.36.09-.32.05-.24.02-.16-.01h-8.22v.82h5.84l.01 2.76.02.36-.05.34-.11.31-.17.29-.25.25-.31.24-.38.2-.44.17-.51.15-.58.13-.64.09-.71.07-.77.04-.84.01-1.27-.04-1.07-.14-.9-.2-.73-.25-.59-.3-.45-.33-.34-.34-.25-.34-.16-.33-.1-.3-.04-.25-.02-.2.01-.13v-5.34l.05-.64.13-.54.21-.46.26-.38.3-.32.33-.24.35-.2.35-.14.33-.1.3-.06.26-.04.21-.02.13-.01h5.84l.69-.05.59-.14.5-.21.41-.28.33-.32.27-.35.2-.36.15-.36.1-.35.07-.32.04-.28.02-.21V6.07h2.09l.14.01zm-6.47 14.25l-.23.33-.08.41.08.41.23.33.33.23.41.08.41-.08.33-.23.23-.33.08-.41-.08-.41-.23-.33-.33-.23-.41-.08-.41.08z","Python","python","Offers a comprehensive CLI for direct interaction with biomedical databases through Python, allowing users to search for articles, clinical trials, and variants via command line.","1mnf3x83k7","namespace",{"_118":119,"_104":123},"owner",{"_120":121},"githubAccount",{"_122":123},"login","genomoncology","readmeMarkdown","redditPostExternalId","repository",{"_128":-5,"_129":130,"_134":135,"_154":155,"_160":161,"_162":-5,"_163":164},"argumentsJsonSchema","githubProject",{"_33":131,"_132":133},"# BioMCP: Biomedical Model Context Protocol\n\nBioMCP is an open source (MIT License) toolkit that empowers AI assistants and\nagents with specialized biomedical knowledge. Built following the Model Context\nProtocol (MCP), it connects AI systems to authoritative biomedical data\nsources, enabling them to answer questions about clinical trials, scientific\nliterature, and genomic variants with precision and depth.\n\n[![▶️ Watch the video](./docs/blog/images/what_is_biomcp_thumbnail.png)](https://www.youtube.com/watch?v=bKxOWrWUUhM)\n\n## MCPHub Certification\n\nBioMCP is certified by [MCPHub](https://mcphub.com/mcp-servers/genomoncology/biomcp). This certification ensures that BioMCP follows best practices for Model Context Protocol implementation and provides reliable biomedical data access.\n\n## Why BioMCP?\n\nWhile Large Language Models have broad general knowledge, they often lack\nspecialized domain-specific information or access to up-to-date resources.\nBioMCP bridges this gap for biomedicine by:\n\n- Providing **structured access** to clinical trials, biomedical literature,\n and genomic variants\n- Enabling **natural language queries** to specialized databases without\n requiring knowledge of their specific syntax\n- Supporting **biomedical research** workflows through a consistent interface\n- Functioning as an **MCP server** for AI assistants and agents\n\n## Biomedical Data Sources\n\nBioMCP integrates with multiple biomedical data sources:\n\n### Literature Sources\n\n- **PubTator3/PubMed** - Peer-reviewed biomedical literature with entity annotations\n- **bioRxiv/medRxiv** - Preprint servers for biology and health sciences\n- **Europe PMC** - Open science platform including preprints\n\n### Clinical & Genomic Sources\n\n- **ClinicalTrials.gov** - Clinical trial registry and results database\n- **NCI Clinical Trials Search API** - National Cancer Institute's curated cancer trials database\n - Advanced search filters (biomarkers, prior therapies, brain metastases)\n - Organization and intervention databases\n - Disease vocabulary with synonyms\n- **BioThings Suite** - Comprehensive biomedical data APIs:\n - **MyVariant.info** - Consolidated genetic variant annotation\n - **MyGene.info** - Real-time gene annotations and information\n - **MyDisease.info** - Disease ontology and synonym information\n - **MyChem.info** - Drug/chemical annotations and properties\n- **TCGA/GDC** - The Cancer Genome Atlas for cancer variant data\n- **1000 Genomes** - Population frequency data via Ensembl\n- **cBioPortal** - Cancer genomics portal with mutation occurrence data\n\n### Regulatory & Safety Sources\n\n- **OpenFDA** - FDA regulatory and safety data:\n - **Drug Adverse Events (FAERS)** - Post-market drug safety reports\n - **Drug Labels (SPL)** - Official prescribing information\n - **Device Events (MAUDE)** - Medical device adverse events, with genomic device filtering\n\n## Available MCP Tools\n\nBioMCP provides 24 specialized tools for biomedical research:\n\n### Core Tools (3)\n\n#### 1. Think Tool (ALWAYS USE FIRST!)\n\n**CRITICAL**: The `think` tool MUST be your first step for ANY biomedical research task.\n\n```python\n# Start analysis with sequential thinking\nthink(\n thought=\"Breaking down the query about BRAF mutations in melanoma...\",\n thoughtNumber=1,\n totalThoughts=3,\n nextThoughtNeeded=True\n)\n```\n\nThe sequential thinking tool helps:\n\n- Break down complex biomedical problems systematically\n- Plan multi-step research approaches\n- Track reasoning progress\n- Ensure comprehensive analysis\n\n#### 2. Search Tool\n\nThe search tool supports two modes:\n\n##### Unified Query Language (Recommended)\n\nUse the `query` parameter with structured field syntax for powerful cross-domain searches:\n\n```python\n# Simple natural language\nsearch(query=\"BRAF melanoma\")\n\n# Field-specific search\nsearch(query=\"gene:BRAF AND trials.condition:melanoma\")\n\n# Complex queries\nsearch(query=\"gene:BRAF AND variants.significance:pathogenic AND articles.date:>2023\")\n\n# Get searchable fields schema\nsearch(get_schema=True)\n\n# Explain how a query is parsed\nsearch(query=\"gene:BRAF\", explain_query=True)\n```\n\n**Supported Fields:**\n\n- **Cross-domain**: `gene:`, `variant:`, `disease:`\n- **Trials**: `trials.condition:`, `trials.phase:`, `trials.status:`, `trials.intervention:`\n- **Articles**: `articles.author:`, `articles.journal:`, `articles.date:`\n- **Variants**: `variants.significance:`, `variants.rsid:`, `variants.frequency:`\n\n##### Domain-Based Search\n\nUse the `domain` parameter with specific filters:\n\n```python\n# Search articles (includes automatic cBioPortal integration)\nsearch(domain=\"article\", genes=[\"BRAF\"], diseases=[\"melanoma\"])\n\n# Search with mutation-specific cBioPortal data\nsearch(domain=\"article\", genes=[\"BRAF\"], keywords=[\"V600E\"])\nsearch(domain=\"article\", genes=[\"SRSF2\"], keywords=[\"F57*\"]) # Wildcard patterns\n\n# Search trials\nsearch(domain=\"trial\", conditions=[\"lung cancer\"], phase=\"3\")\n\n# Search variants\nsearch(domain=\"variant\", gene=\"TP53\", significance=\"pathogenic\")\n```\n\n**Note**: When searching articles with a gene parameter, cBioPortal data is automatically included:\n\n- Gene-level summaries show mutation frequency across cancer studies\n- Mutation-specific searches (e.g., \"V600E\") show study-level occurrence data\n- Cancer types are dynamically resolved from cBioPortal API\n\n#### 3. Fetch Tool\n\nRetrieve full details for a single article, trial, or variant:\n\n```python\n# Fetch article details (supports both PMID and DOI)\nfetch(domain=\"article\", id=\"34567890\") # PMID\nfetch(domain=\"article\", id=\"10.1101/2024.01.20.23288905\") # DOI\n\n# Fetch trial with all sections\nfetch(domain=\"trial\", id=\"NCT04280705\", detail=\"all\")\n\n# Fetch variant details\nfetch(domain=\"variant\", id=\"rs113488022\")\n```\n\n**Domain-specific options:**\n\n- **Articles**: `detail=\"full\"` retrieves full text if available\n- **Trials**: `detail` can be \"protocol\", \"locations\", \"outcomes\", \"references\", or \"all\"\n- **Variants**: Always returns full details\n\n### Individual Tools (21)\n\nFor users who prefer direct access to specific functionality, BioMCP also provides 21 individual tools:\n\n#### Article Tools (2)\n\n- **article_searcher**: Search PubMed/PubTator3 and preprints\n- **article_getter**: Fetch detailed article information (supports PMID and DOI)\n\n#### Trial Tools (5)\n\n- **trial_searcher**: Search ClinicalTrials.gov or NCI CTS API (via source parameter)\n- **trial_getter**: Fetch all trial details from either source\n- **trial_protocol_getter**: Fetch protocol information only (ClinicalTrials.gov)\n- **trial_references_getter**: Fetch trial publications (ClinicalTrials.gov)\n- **trial_outcomes_getter**: Fetch outcome measures and results (ClinicalTrials.gov)\n- **trial_locations_getter**: Fetch site locations and contacts (ClinicalTrials.gov)\n\n#### Variant Tools (2)\n\n- **variant_searcher**: Search MyVariant.info database\n- **variant_getter**: Fetch comprehensive variant details\n\n#### NCI-Specific Tools (6)\n\n- **nci_organization_searcher**: Search NCI's organization database\n- **nci_organization_getter**: Get organization details by ID\n- **nci_intervention_searcher**: Search NCI's intervention database (drugs, devices, procedures)\n- **nci_intervention_getter**: Get intervention details by ID\n- **nci_biomarker_searcher**: Search biomarkers used in trial eligibility criteria\n- **nci_disease_searcher**: Search NCI's controlled vocabulary of cancer conditions\n\n#### Gene, Disease & Drug Tools (3)\n\n- **gene_getter**: Get real-time gene information from MyGene.info\n- **disease_getter**: Get disease definitions and synonyms from MyDisease.info\n- **drug_getter**: Get drug/chemical information from MyChem.info\n\n**Note**: All individual tools that search by gene automatically include cBioPortal summaries when the `include_cbioportal` parameter is True (default). Trial searches can expand disease conditions with synonyms when `expand_synonyms` is True (default).\n\n## Quick Start\n\n### For Claude Desktop Users\n\n1. **Install `uv`** if you don't have it (recommended):\n\n ```bash\n # MacOS\n brew install uv\n\n # Windows/Linux\n pip install uv\n ```\n\n2. **Configure Claude Desktop**:\n - Open Claude Desktop settings\n - Navigate to Developer section\n - Click \"Edit Config\" and add:\n ```json\n {\n \"mcpServers\": {\n \"biomcp\": {\n \"command\": \"uv\",\n \"args\": [\"run\", \"--with\", \"biomcp-python\", \"biomcp\", \"run\"]\n }\n }\n }\n ```\n - Restart Claude Desktop and start chatting about biomedical topics!\n\n### Python Package Installation\n\n```bash\n# Using pip\npip install biomcp-python\n\n# Using uv (recommended for faster installation)\nuv pip install biomcp-python\n\n# Run directly without installation\nuv run --with biomcp-python biomcp trial search --condition \"lung cancer\"\n```\n\n## Configuration\n\n### Environment Variables\n\nBioMCP supports optional environment variables for enhanced functionality:\n\n```bash\n# cBioPortal API authentication (optional)\nexport CBIO_TOKEN=\"your-api-token\" # For authenticated access\nexport CBIO_BASE_URL=\"https://www.cbioportal.org/api\" # Custom API endpoint\n\n# Performance tuning\nexport BIOMCP_USE_CONNECTION_POOL=\"true\" # Enable HTTP connection pooling (default: true)\nexport BIOMCP_METRICS_ENABLED=\"false\" # Enable performance metrics (default: false)\n```\n\n## Running BioMCP Server\n\nBioMCP supports multiple transport protocols to suit different deployment scenarios:\n\n### Local Development (STDIO)\n\nFor direct integration with Claude Desktop or local MCP clients:\n\n```bash\n# Default STDIO mode for local development\nbiomcp run\n\n# Or explicitly specify STDIO\nbiomcp run --mode stdio\n```\n\n### HTTP Server Mode\n\nBioMCP supports multiple HTTP transport protocols:\n\n#### Legacy SSE Transport (Worker Mode)\n\nFor backward compatibility with existing SSE clients:\n\n```bash\nbiomcp run --mode worker\n# Server available at http://localhost:8000/sse\n```\n\n#### Streamable HTTP Transport (Recommended)\n\nThe new MCP-compliant Streamable HTTP transport provides optimal performance and standards compliance:\n\n```bash\nbiomcp run --mode streamable_http\n\n# Custom host and port\nbiomcp run --mode streamable_http --host 127.0.0.1 --port 8080\n```\n\nFeatures of Streamable HTTP transport:\n\n- Single `/mcp` endpoint for all operations\n- Dynamic response mode (JSON for quick operations, SSE for long-running)\n- Session management support (future)\n- Full MCP specification compliance (2025-03-26)\n- Better scalability for cloud deployments\n\n### Deployment Options\n\n#### Docker\n\n```bash\n# Build the Docker image locally\ndocker build -t biomcp:latest .\n\n# Run the container\ndocker run -p 8000:8000 biomcp:latest biomcp run --mode streamable_http\n```\n\n#### Cloudflare Workers\n\nThe worker mode can be deployed to Cloudflare Workers for global edge deployment.\n\nNote: All APIs work without authentication, but tokens may provide higher rate limits.\n\n## Command Line Interface\n\nBioMCP provides a comprehensive CLI for direct database interaction:\n\n```bash\n# Get help\nbiomcp --help\n\n# Run the MCP server\nbiomcp run\n\n# Article search examples\nbiomcp article search --gene BRAF --disease Melanoma # Includes preprints by default\nbiomcp article search --gene BRAF --no-preprints # Exclude preprints\nbiomcp article get 21717063 --full\n\n# Clinical trial examples\nbiomcp trial search --condition \"Lung Cancer\" --phase PHASE3\nbiomcp trial search --condition melanoma --source nci --api-key YOUR_KEY # Use NCI API\nbiomcp trial get NCT04280705 Protocol\nbiomcp trial get NCT04280705 --source nci --api-key YOUR_KEY # Get from NCI\n\n# Variant examples with external annotations\nbiomcp variant search --gene TP53 --significance pathogenic\nbiomcp variant get rs113488022 # Includes TCGA, 1000 Genomes, and cBioPortal data by default\nbiomcp variant get rs113488022 --no-external # Core annotations only\n\n# NCI-specific examples (requires NCI API key)\nbiomcp organization search \"MD Anderson\" --api-key YOUR_KEY\nbiomcp organization get ORG123456 --api-key YOUR_KEY\nbiomcp intervention search pembrolizumab --api-key YOUR_KEY\nbiomcp intervention search --type Device --api-key YOUR_KEY\nbiomcp biomarker search \"PD-L1\" --api-key YOUR_KEY\nbiomcp disease search melanoma --source nci --api-key YOUR_KEY\n```\n\n## Testing & Verification\n\nTest your BioMCP setup with the MCP Inspector:\n\n```bash\nnpx @modelcontextprotocol/inspector uv run --with biomcp-python biomcp run\n```\n\nThis opens a web interface where you can explore and test all available tools.\n\n## Enterprise Version: OncoMCP\n\nOncoMCP extends BioMCP with GenomOncology's enterprise-grade precision oncology\nplatform (POP), providing:\n\n- **HIPAA-Compliant Deployment**: Secure on-premise options\n- **Real-Time Trial Matching**: Up-to-date status and arm-level matching\n- **Healthcare Integration**: Seamless EHR and data warehouse connectivity\n- **Curated Knowledge Base**: 15,000+ trials and FDA approvals\n- **Sophisticated Patient Matching**: Using integrated clinical and molecular\n profiles\n- **Advanced NLP**: Structured extraction from unstructured text\n- **Comprehensive Biomarker Processing**: Mutation and rule processing\n\nLearn more: [GenomOncology](https://genomoncology.com/)\n\n## MCP Registries\n\n[![smithery badge](https://smithery.ai/badge/@genomoncology/biomcp)](https://smithery.ai/server/@genomoncology/biomcp)\n\n\n

\n\n\n## Example Use Cases\n\n### Gene Information Retrieval\n\n```python\n# Get comprehensive gene information\ngene_getter(gene_id_or_symbol=\"TP53\")\n# Returns: Official name, summary, aliases, links to databases\n```\n\n### Disease Synonym Expansion\n\n```python\n# Get disease information with synonyms\ndisease_getter(disease_id_or_name=\"GIST\")\n# Returns: \"gastrointestinal stromal tumor\" and other synonyms\n\n# Search trials with automatic synonym expansion\ntrial_searcher(conditions=[\"GIST\"], expand_synonyms=True)\n# Searches for: GIST OR \"gastrointestinal stromal tumor\" OR \"GI stromal tumor\"\n```\n\n### Integrated Biomedical Research\n\n```python\n# 1. Always start with thinking\nthink(thought=\"Analyzing BRAF V600E in melanoma treatment\", thoughtNumber=1)\n\n# 2. Get gene context\ngene_getter(\"BRAF\")\n\n# 3. Search for pathogenic variants\nvariant_searcher(gene=\"BRAF\", hgvsp=\"V600E\", significance=\"pathogenic\")\n\n# 4. Find relevant clinical trials with disease expansion\ntrial_searcher(conditions=[\"melanoma\"], interventions=[\"BRAF inhibitor\"])\n```\n\n## Documentation\n\nFor comprehensive documentation, visit [https://biomcp.org](https://biomcp.org)\n\n### Developer Guides\n\n- [HTTP Client Guide](./docs/http-client-guide.md) - Using the centralized HTTP client\n- [Migration Examples](./docs/migration-examples.md) - Migrating from direct HTTP usage\n- [Error Handling Guide](./docs/error-handling.md) - Comprehensive error handling patterns\n- [Integration Testing Guide](./docs/integration-testing.md) - Best practices for reliable integration tests\n- [Third-Party Endpoints](./THIRD_PARTY_ENDPOINTS.md) - Complete list of external APIs used\n- [Testing Guide](./docs/development/testing.md) - Running tests and understanding test categories\n\n## Development\n\n### Running Tests\n\n```bash\n# Run all tests (including integration tests)\nmake test\n\n# Run only unit tests (excluding integration tests)\nuv run python -m pytest tests -m \"not integration\"\n\n# Run only integration tests\nuv run python -m pytest tests -m \"integration\"\n```\n\n**Note**: Integration tests make real API calls and may fail due to network issues or rate limiting.\nIn CI/CD, integration tests are run separately and allowed to fail without blocking the build.\n\n## BioMCP Examples Repo\n\nLooking to see BioMCP in action?\n\nCheck out the companion repository:\n👉 **[biomcp-examples](https://github.com/genomoncology/biomcp-examples)**\n\nIt contains real prompts, AI-generated research briefs, and evaluation runs across different models.\nUse it to explore capabilities, compare outputs, or benchmark your own setup.\n\nHave a cool example of your own?\n**We’d love for you to contribute!** Just fork the repo and submit a PR with your experiment.\n\n## License\n\nThis project is licensed under the MIT License.\n","url","https://github.com/genomoncology/biomcp","githubRepository",{"_136":137,"_138":139,"_140":141,"_16":142,"_143":112,"_31":144,"_118":145,"_147":148,"_149":150,"_152":153},"defaultBranch","main","fullName","genomoncology/biomcp","githubTreeId",1485049,9173,"language","biomcp",{"_122":123,"_107":146},"x8qxXq2eX0","recentStargazers",3,"spdxLicense",{"_31":151},"MIT License","stargazers",327,"glamaJson",{"_156":157},"maintainers",[158,159],"imaurer","jyeakley","inspectable",true,"npmPackage","supportedPlatforms",[165,166,167],"MACOS","WINDOWS","LINUX","scores",{"_170":171,"_172":171,"_173":171},"license",100,"quality","security","simpleIcon","toolCount",35,"va4wpvdyc3","updatedAt","2025-10-03T18:55:42.629843Z","meta",[182,185,187,194,197,199,202,205,208,211,214,216,219,222,224,226,228,230,232],{"_183":184},"title","BioMCP | Glama",{"_186":40,"_31":29},"content",{"_188":189,"_190":191,"_192":193},"href","https://glama.ai/mcp/servers/@genomoncology/biomcp","rel","canonical","tagName","link",{"_186":32,"_195":196},"property","og:title",{"_186":40,"_195":198},"og:description",{"_186":200,"_195":201},"https://glama.ai/logo.png","og:logo",{"_186":203,"_195":204},"https://glama.ai/mcp/servers/@genomoncology/biomcp/og-image","og:image",{"_186":206,"_195":207},"630","og:image:height",{"_186":209,"_195":210},"1200","og:image:width",{"_186":212,"_195":213},"website","og:type",{"_186":189,"_195":215},"og:url",{"_186":217,"_195":218},"Glama – MCP Hosting Platform","og:site_name",{"_186":220,"_31":221},"summary_large_image","twitter:card",{"_186":217,"_31":223},"twitter:title",{"_186":40,"_31":225},"twitter:description",{"_186":203,"_31":227},"twitter:image",{"_186":206,"_31":229},"twitter:image:height",{"_186":209,"_31":231},"twitter:image:width",{"_233":234},"script:ld+json",{"_235":236,"_237":238},"@context","http://schema.org","@graph",[239],{"_235":240,"_241":242,"_243":244},"https://schema.org","@type","BreadcrumbList","itemListElement",[245,252,256,259],{"_241":246,"_247":248,"_31":249,"_250":251},"ListItem","item","/","Glama","position",1,{"_241":246,"_247":253,"_31":254,"_250":255},"/mcp","MCP",2,{"_241":246,"_247":257,"_31":258,"_250":148},"/mcp/servers","Servers",{"_241":246,"_247":260,"_31":32,"_250":261},"/mcp/servers/@genomoncology/biomcp",4,"visitorIsMaintainer",false,"routes/_public/mcp/servers/~namespace/~slug/_pages/_explorer/blob/~commit/~/_route",{"_266":267,"_27":273,"_292":293,"_180":307,"_304":331,"_332":333},"blob",{"_186":268,"_269":270,"_271":272},"Feature: Autocomplete\n\n Scenario Outline: Valid autocomplete concept and query\n Given I have a valid concept \"\" and a valid query \"\"\n When I call the Pubtator Autocomplete API\n Then the response entity_id should be \"\"\n And the response concept should be \"\"\n\n Examples:\n | concept | query | expected_id |\n | gene | her2 | @GENE_ERBB2 |\n | variant | BRAF V600E | @VARIANT_p.V600E_BRAF_human |\n | disease | lung adenocarcinoma | @DISEASE_Adenocarcinoma_of_Lung |\n | chemical | Caffeine | @CHEMICAL_Caffeine |\n\n Scenario Outline: Invalid autocomplete query\n Given I have a valid concept \"\" and an invalid query \"\"\n When I call the Pubtator Autocomplete API\n Then the response should be empty\n\n Examples:\n | concept | query |\n | gene | iPhone |\n | variant | iPhone |\n | disease | iPhone |\n | chemical | iPhone |\n","fileName","autocomplete.feature","size",1061,{"_81":82,"_13":274,"_93":40,"_94":40,"_16":95,"_96":275,"_31":32,"_116":280,"_124":-5,"_125":-5,"_126":283,"_168":291,"_174":-5,"_104":144,"_175":176,"_107":177,"_178":179},[84,85,86,87,88,89,90,91,92],[276,278],{"_99":277,"_29":106,"_107":108},{"_101":102,"_31":103,"_104":105},{"_99":279,"_29":114,"_107":115},{"_101":111,"_31":112,"_104":113},{"_118":281,"_104":123},{"_120":282},{"_122":123},{"_128":-5,"_129":284,"_134":285,"_154":288,"_160":161,"_162":-5,"_163":290},{"_33":131,"_132":133},{"_136":137,"_138":139,"_140":141,"_16":142,"_143":112,"_31":144,"_118":286,"_147":148,"_149":287,"_152":153},{"_122":123,"_107":146},{"_31":151},{"_156":289},[158,159],[165,166,167],{"_170":171,"_172":171,"_173":171},"member",{"_294":295,"_298":299,"_300":301,"_140":302,"_303":263,"_304":270,"_305":161,"_271":272,"_306":266},"commit",{"_296":297},"sha","388dc5d8142b6f653df17590d9c72f134b80c1e0","contentStoredObjectPointerId",11023865,"fullPath","tests/bdd/search_articles/autocomplete.feature",1485027,"indexable","path","safeToDisplayAsText","type",[308,310,312,314,316,317,318,320,321,322,323,324,325,326,327,328,329,330],{"_183":309},"autocomplete.feature | BioMCP | Glama",{"_186":311,"_31":29},"Enterprise-grade security, privacy, and features like agents, MCP, prompt templates, and more.",{"_188":313,"_190":191,"_192":193},"https://glama.ai/mcp/servers/@genomoncology/biomcp/blob/388dc5d8142b6f653df17590d9c72f134b80c1e0/tests/bdd/search_articles",{"_186":315,"_195":196},"autocomplete.feature | BioMCP",{"_186":311,"_195":198},{"_186":200,"_195":201},{"_186":319,"_195":204},"https://glama.ai/generated-images/og?title=autocomplete.feature+%7C+BioMCP",{"_186":206,"_195":207},{"_186":209,"_195":210},{"_186":212,"_195":213},{"_186":313,"_195":215},{"_186":217,"_195":218},{"_186":220,"_31":221},{"_186":217,"_31":223},{"_186":311,"_31":225},{"_186":319,"_31":227},{"_186":206,"_31":229},{"_186":209,"_31":231},"tests/bdd/search_articles","treeMembers",[334,336,340,344],{"_294":335,"_300":301,"_304":270,"_306":266},{"_296":297},{"_294":337,"_300":338,"_304":339,"_306":266},{"_296":297},"tests/bdd/search_articles/search.feature","search.feature",{"_294":341,"_300":342,"_304":343,"_306":266},{"_296":297},"tests/bdd/search_articles/test_autocomplete.py","test_autocomplete.py",{"_294":345,"_300":346,"_304":347,"_306":266},{"_296":297},"tests/bdd/search_articles/test_search.py","test_search.py","actionData","errors"]

MCP directory API