[{"_1":2,"_376":-5,"_377":-5},"loaderData",{"_3":4,"_29":30,"_279":280},"root",{"_5":6,"_14":15},"toasts",[7],{"_8":9,"_10":11,"_12":13},"uid","0199c385-9ff9-763c-9bba-9ab10c9b1aeb","title","This MCP server does not support inspection","type","error","visitor",{"_16":17,"_18":19},"viewportWidth",1280,"visitorSession",{"_20":21,"_23":24,"_25":-5,"_26":27,"_28":-5},"attributes",[22],"bot","id",17782814,"membership","referenceId","0199bf5f-ca2b-7322-a5a4-336805538485","userAccount","routes/_public/mcp/servers/~namespace/~slug/_pages/_layout",{"_31":32,"_87":-5,"_88":45,"_34":89,"_196":197,"_277":278},"availableLocales",[33,45,52,59,66,73,80],{"_34":35,"_38":42,"_43":44},"mcpServer",{"_36":37,"_38":39,"_40":41},"description","Ein Model Context Protocol-Server, der KI-Clients die Interaktion mit der Natural Language Query API von BlazeSQL ermöglicht und so Abfragen in natürlicher Sprache an SQL-Datenbanken zulässt.","name","BlazeSQL MCP-Server","readme","# BlazeSQL MCP-Server\n\nDieses Projekt implementiert einen Model Context Protocol (MCP)-Server mit dem `@modelcontextprotocol/sdk` der als Proxy für die BlazeSQL Natural Language Query API fungiert. Es ermöglicht MCP-kompatiblen Clients (wie Cursor, Claude 3 mit Tool-Nutzung, dem MCP Inspector usw.) die Interaktion mit BlazeSQL in natürlicher Sprache.\n\n## Merkmale\n\n* Erstellt mit der modernen `McpServer` Hilfsklasse aus dem MCP SDK.\n* Stellt die BlazeSQL Natural Language Query API als MCP-Tool namens `blazesql_query` bereit.\n* Verwendet `zod` zur robusten Validierung der Tool-Eingabeparameter.\n* Behandelt die API-Schlüsselauthentifizierung sicher über Umgebungsvariablen.\n* Kommuniziert mit Clients über den standardmäßigen MCP-Stdio-Transport.\n\n## Workflow-Diagramm\n\nDieses Diagramm zeigt die Abfolge der Interaktionen, wenn ein Client das Tool `blazesql_query` verwendet (Hinweis: Die interne Serverlogik verwendet jetzt `McpServer` , was die Toolregistrierung im Vergleich zu den im Diagramm gezeigten Low-Level-Handlern vereinfacht):\n\n```mermaid\nsequenceDiagram\n participant Client as MCP Client (e.g., Cursor)\n participant Server as BlazeSQL MCP Server (index.ts)\n participant Env as Environment (.env)\n participant BlazeAPI as BlazeSQL API\n\n Client->>Server: ListTools Request (via stdio)\n Server-->>Client: ListTools Response (tools: [blazesql_query]) (via stdio)\n\n Client->>Server: CallTool Request (blazesql_query, db_id, nl_request) (via stdio)\n Server->>Env: Read BLAZE_API_KEY\n Env-->>Server: BLAZE_API_KEY\n Server->>BlazeAPI: POST /natural_language_query_api (apiKey, db_id, nl_request)\n BlazeAPI->>BlazeAPI: Process Query (NL->SQL, Execute)\n BlazeAPI-->>Server: HTTPS Response (JSON: agent_response, query, data_result OR error)\n Server->>Server: Format Response (Agent response, SQL, and data into single text block)\n Server-->>Client: CallTool Response (content: [{type: text, text: formattedMarkdown}]) (via stdio)\n\n```\n\n## Voraussetzungen\n\n* [Node.js](https://nodejs.org/) (LTS-Version empfohlen)\n* [Garn](https://yarnpkg.com/) (Klassisch oder Beere)\n* Ein BlazeSQL-Konto mit einem API-Schlüssel (für die API ist ein Team Advanced-Abonnement erforderlich).\n* Mindestens eine Datenbankverbindung ist in Ihrem BlazeSQL-Konto konfiguriert.\n* BlazeSQL Natural Language Query API-Dokumentation: \n\n## Aufstellen\n\n1. **Klonen Sie das Repository:**\n\n ```bash\n git clone \n cd blaze-sql-mcp-server\n ```\n\n2. **Abhängigkeiten installieren:**\n\n ```bash\n yarn install\n ```\n\n Dadurch werden alle erforderlichen Abhängigkeiten installiert, einschließlich `@modelcontextprotocol/sdk` , `dotenv` und `zod` .\n\n3. **Umgebungsvariablen konfigurieren:**\n\n * Kopieren Sie die Beispielumgebungsdatei:\n\n ```bash\n cp .env.sample .env\n ```\n\n * Bearbeiten Sie die `.env` Datei:\n\n ```dotenv\n # .env\n BLAZE_API_KEY=YOUR_BLAZESQL_API_KEY_HERE\n ```\n\n Ersetzen Sie `YOUR_BLAZESQL_API_KEY_HERE` durch Ihren tatsächlichen API-Schlüssel, den Sie aus Ihren BlazeSQL-Kontoeinstellungen erhalten haben.\n\n## Ausführen des Servers\n\n1. **Erstellen Sie den Server:** Kompilieren Sie den TypeScript-Code in JavaScript:\n\n ```bash\n yarn build\n ```\n\n2. **Führen Sie den Server aus:** Führen Sie den kompilierten Code aus:\n\n ```bash\n node build/index.js\n ```\n\n Der Server startet und protokolliert Meldungen in `stderr` (möglicherweise wird „API-Schlüssel erfolgreich geladen …“ usw. angezeigt). Er wartet nun über die Standard-Ein-/Ausgabe (stdio) auf eine MCP-Clientverbindung.\n\n## Verbinden eines MCP-Clients\n\nDieser Server verwendet den **stdio-** Transportmechanismus.\n\n### Verwenden des MCP Inspector (zum Testen empfohlen)\n\n1. Stellen Sie sicher, dass der Server **nicht** bereits separat ausgeführt wird.\n\n2. Führen Sie den Inspector aus und weisen Sie ihn an, Ihren Server zu starten:\n\n ```bash\n npx @modelcontextprotocol/inspector node build/index.js\n ```\n\n3. Die Inspector-Benutzeroberfläche wird gestartet und stellt automatisch eine Verbindung zu Ihrem Server her.\n\n4. Navigieren Sie zur Registerkarte „Tools“, um mit dem Tool `blazesql_query` zu interagieren.\n\n### Verwendung integrierter Clients (Cursor, Claude 3 usw.)\n\n1. **Starten Sie den Server** in einem Terminal:\n\n ```bash\n node build/index.js\n ```\n\n2. **Konfigurieren Sie den Client:** In den Einstellungen Ihres MCP-Clients müssen Sie eine benutzerdefinierte Serverkonfiguration hinzufügen.\n\n * **Transport:** Wählen Sie `stdio` aus.\n * **Befehl:** Geben Sie den genauen Befehl zum Ausführen des Servers an. Sie müssen den **absoluten Pfad** zum Knoten und den **absoluten Pfad** zur Datei `build/index.js` angeben.\n * Beispiel (macOS/Linux – Pfade nach Bedarf anpassen): `/usr/local/bin/node /Users/your_username/path/to/blaze-sql-mcp-server/build/index.js`\n * Sie können den Pfad zum Knoten mithilfe der Angabe `which node` in Ihrem Terminal finden.\n * Den Pfad zum Projekt finden Sie mit `pwd` im Projektverzeichnis.\n * Speichern Sie die Konfiguration.\n\n3. Der Client sollte jetzt in der Lage sein, eine Verbindung zu Ihrem lokal laufenden Server herzustellen und dessen Tools aufzulisten/verwenden.\n\n## Verwenden des `blazesql_query` -Tools\n\nSobald die Verbindung hergestellt ist, kann der Client das Tool `blazesql_query` aufrufen.\n\n* **Toolname:** `blazesql_query`\n\n* **Argumente:**\n\n * `db_id` (Zeichenfolge, erforderlich): Die ID der Zieldatenbankverbindung in Ihrem BlazeSQL-Konto. Sie finden diese ID in der BlazeSQL-Webanwendung beim Verwalten Ihrer Datenbankverbindungen.\n * `natural_language_request` (Zeichenfolge, erforderlich): Die Abfrage, die Sie ausführen möchten, in einfachem Englisch (z. B. „Zeige mir die Gesamtzahl der Benutzer“). *(Die Eingabe wird mit `zod` validiert.)*\n\n* **Beispielaufruf (zur Veranschaulichung wird `mcp test` verwendet):**\n\n ```bash\n call-tool blazesql_query --db_id \"db_your_actual_db_id\" --natural_language_request \"What were the total sales last month?\"\n ```\n\n* **Ausgabe:** Bei Erfolg gibt das Tool einen einzelnen `text` zurück, der Folgendes enthält:\n\n * Die Antwort des BlazeSQL-Agenten in natürlicher Sprache.\n * Die generierte SQL-Abfrage innerhalb eines Markdown-Code-Zauns ( `sql ...` ).\n * Die Datenergebnisse werden als JSON innerhalb eines Markdown-Code-Zauns ( `json ...` ) formatiert.\n\n Beispielhafte Struktur innerhalb des `text` :\n\n ````markdown\n **Agent Response:**\n The total sales last month were $12345.67.\n\n **Generated SQL:**\n ```sql\n SELECT sum(sales_amount) FROM sales WHERE sale_date >= date('now', '-1 month');\n ````\n\n **Datenergebnis (JSON):**\n\n ```json\n [\n {\n \"sum(sales_amount)\": 12345.67\n }\n ]\n ```\n\n ```\n\n If unsuccessful, it returns a `text` content block containing the error message from the BlazeSQL API and marks the response as an error (`isError: true`).\n ```","German","tag","de-DE",{"_34":46,"_38":50,"_43":51},{"_36":47,"_38":48,"_40":49},"A Model Context Protocol server that enables AI clients to interact with BlazeSQL's Natural Language Query API, allowing natural language queries against SQL databases.","BlazeSQL MCP Server","# BlazeSQL MCP Server\n\nThis project implements a Model Context Protocol (MCP) server using the `@modelcontextprotocol/sdk` that acts as a proxy to the BlazeSQL Natural Language Query API. It allows MCP-compatible clients (like Cursor, Claude 3 with tool use, the MCP Inspector, etc.) to interact with BlazeSQL using natural language.\n\n## Features\n\n* Built using the modern `McpServer` helper class from the MCP SDK.\n* Exposes the BlazeSQL Natural Language Query API as an MCP tool named `blazesql_query`.\n* Uses `zod` for robust validation of tool input parameters.\n* Handles API key authentication securely via environment variables.\n* Communicates with clients using the standard MCP stdio transport.\n\n## Workflow Diagram\n\nThis diagram shows the sequence of interactions when a client uses the `blazesql_query` tool (Note: The internal server logic now uses `McpServer` which simplifies tool registration compared to the low-level handlers shown in the diagram):\n\n```mermaid\nsequenceDiagram\n participant Client as MCP Client (e.g., Cursor)\n participant Server as BlazeSQL MCP Server (index.ts)\n participant Env as Environment (.env)\n participant BlazeAPI as BlazeSQL API\n\n Client->>Server: ListTools Request (via stdio)\n Server-->>Client: ListTools Response (tools: [blazesql_query]) (via stdio)\n\n Client->>Server: CallTool Request (blazesql_query, db_id, nl_request) (via stdio)\n Server->>Env: Read BLAZE_API_KEY\n Env-->>Server: BLAZE_API_KEY\n Server->>BlazeAPI: POST /natural_language_query_api (apiKey, db_id, nl_request)\n BlazeAPI->>BlazeAPI: Process Query (NL->SQL, Execute)\n BlazeAPI-->>Server: HTTPS Response (JSON: agent_response, query, data_result OR error)\n Server->>Server: Format Response (Agent response, SQL, and data into single text block)\n Server-->>Client: CallTool Response (content: [{type: text, text: formattedMarkdown}]) (via stdio)\n\n```\n\n## Prerequisites\n\n* [Node.js](https://nodejs.org/) (LTS version recommended)\n* [Yarn](https://yarnpkg.com/) (Classic or Berry)\n* A BlazeSQL account with an API Key (Team Advanced subscription required for the API).\n* At least one database connection configured in your BlazeSQL account.\n* BlazeSQL Natural Language Query API Documentation: [https://help.blazesql.com/en/article/natural-language-query-api-1fgx4au/](https://help.blazesql.com/en/article/natural-language-query-api-1fgx4au/)\n\n## Setup\n\n1. **Clone the Repository:**\n ```bash\n git clone \n cd blaze-sql-mcp-server\n ```\n\n2. **Install Dependencies:**\n ```bash\n yarn install\n ```\n This will install all necessary dependencies, including the `@modelcontextprotocol/sdk`, `dotenv`, and `zod`.\n\n3. **Configure Environment Variables:**\n * Copy the example environment file:\n ```bash\n cp .env.sample .env\n ```\n * Edit the `.env` file:\n ```dotenv\n # .env\n BLAZE_API_KEY=YOUR_BLAZESQL_API_KEY_HERE\n ```\n Replace `YOUR_BLAZESQL_API_KEY_HERE` with your actual API key obtained from your BlazeSQL account settings.\n\n## Running the Server\n\n1. **Build the Server:**\n Compile the TypeScript code to JavaScript:\n ```bash\n yarn build\n ```\n\n2. **Run the Server:**\n Execute the compiled code:\n ```bash\n node build/index.js\n ```\n The server will start and log messages to `stderr` (you might see \"API Key loaded successfully...\" etc.). It is now listening for an MCP client connection via standard input/output (stdio).\n\n## Connecting an MCP Client\n\nThis server uses the **stdio** transport mechanism.\n\n### Using the MCP Inspector (Recommended for Testing)\n\n1. Make sure the server is **not** already running separately.\n2. Run the Inspector, telling it to launch your server:\n ```bash\n npx @modelcontextprotocol/inspector node build/index.js\n ```\n3. The Inspector UI will launch, automatically connecting to your server.\n4. Navigate to the \"Tools\" tab to interact with the `blazesql_query` tool.\n\n### Using Integrated Clients (Cursor, Claude 3, etc.)\n\n1. **Start the server** in a terminal:\n ```bash\n node build/index.js\n ```\n2. **Configure the client:** In your MCP client's settings, you need to add a custom server configuration.\n * **Transport:** Select `stdio`.\n * **Command:** Specify the exact command used to run the server. You need to provide the **absolute path** to node and the **absolute path** to the `build/index.js` file.\n * Example (macOS/Linux - adjust paths as needed):\n `/usr/local/bin/node /Users/your_username/path/to/blaze-sql-mcp-server/build/index.js`\n * You can find the path to node using `which node` in your terminal.\n * You can find the path to the project using `pwd` inside the project directory.\n * Save the configuration.\n3. The client should now be able to connect to your locally running server and list/use its tools.\n\n## Using the `blazesql_query` Tool\n\nOnce connected, the client can call the `blazesql_query` tool.\n\n* **Tool Name:** `blazesql_query`\n* **Arguments:**\n * `db_id` (string, required): The ID of the target database connection in your BlazeSQL account. You can find this ID in the BlazeSQL web application when managing your database connections.\n * `natural_language_request` (string, required): The query you want to execute, written in plain English (e.g., \"show me the total number of users\").\n *(Input is validated using `zod`)*\n\n* **Example Call (using `mcp test` syntax for illustration):**\n ```bash\n call-tool blazesql_query --db_id \"db_your_actual_db_id\" --natural_language_request \"What were the total sales last month?\"\n ```\n\n* **Output:**\n If successful, the tool returns a single `text` content block containing:\n * The natural language response from the BlazeSQL agent.\n * The generated SQL query within a Markdown code fence (```sql ... ```).\n * The data results formatted as JSON within a Markdown code fence (```json ... ```).\n\n Example structure within the `text` block:\n ```markdown\n **Agent Response:**\n The total sales last month were $12345.67.\n\n **Generated SQL:**\n ```sql\n SELECT sum(sales_amount) FROM sales WHERE sale_date >= date('now', '-1 month');\n ```\n\n **Data Result (JSON):**\n ```json\n [\n {\n \"sum(sales_amount)\": 12345.67\n }\n ]\n ```\n ```\n\n If unsuccessful, it returns a `text` content block containing the error message from the BlazeSQL API and marks the response as an error (`isError: true`).\n","English","en-US",{"_34":53,"_38":57,"_43":58},{"_36":54,"_38":55,"_40":56},"Un servidor de protocolo de contexto de modelo que permite a los clientes de IA interactuar con la API de consulta de lenguaje natural de BlazeSQL, lo que permite realizar consultas en lenguaje natural en bases de datos SQL.","Servidor MCP BlazeSQL","# Servidor MCP BlazeSQL\n\nEste proyecto implementa un servidor de Protocolo de Contexto de Modelo (MCP) mediante `@modelcontextprotocol/sdk` que actúa como proxy para la API de Consultas en Lenguaje Natural de BlazeSQL. Permite que los clientes compatibles con MCP (como Cursor, Claude 3 con uso de herramientas, el Inspector de MCP, etc.) interactúen con BlazeSQL mediante lenguaje natural.\n\n## Características\n\n* Construido utilizando la clase auxiliar `McpServer` moderna del SDK de MCP.\n* Expone la API de consulta de lenguaje natural BlazeSQL como una herramienta MCP llamada `blazesql_query` .\n* Utiliza `zod` para una validación robusta de los parámetros de entrada de la herramienta.\n* Maneja la autenticación de clave API de forma segura a través de variables de entorno.\n* Se comunica con los clientes utilizando el transporte stdio MCP estándar.\n\n## Diagrama de flujo de trabajo\n\nEste diagrama muestra la secuencia de interacciones cuando un cliente usa la herramienta `blazesql_query` (Nota: la lógica del servidor interno ahora usa `McpServer` , lo que simplifica el registro de la herramienta en comparación con los controladores de bajo nivel que se muestran en el diagrama):\n\n```mermaid\nsequenceDiagram\n participant Client as MCP Client (e.g., Cursor)\n participant Server as BlazeSQL MCP Server (index.ts)\n participant Env as Environment (.env)\n participant BlazeAPI as BlazeSQL API\n\n Client->>Server: ListTools Request (via stdio)\n Server-->>Client: ListTools Response (tools: [blazesql_query]) (via stdio)\n\n Client->>Server: CallTool Request (blazesql_query, db_id, nl_request) (via stdio)\n Server->>Env: Read BLAZE_API_KEY\n Env-->>Server: BLAZE_API_KEY\n Server->>BlazeAPI: POST /natural_language_query_api (apiKey, db_id, nl_request)\n BlazeAPI->>BlazeAPI: Process Query (NL->SQL, Execute)\n BlazeAPI-->>Server: HTTPS Response (JSON: agent_response, query, data_result OR error)\n Server->>Server: Format Response (Agent response, SQL, and data into single text block)\n Server-->>Client: CallTool Response (content: [{type: text, text: formattedMarkdown}]) (via stdio)\n\n```\n\n## Prerrequisitos\n\n* [Node.js](https://nodejs.org/) (versión LTS recomendada)\n* [Hilo](https://yarnpkg.com/) (clásico o baya)\n* Una cuenta BlazeSQL con una clave API (se requiere suscripción a Team Advanced para la API).\n* Al menos una conexión de base de datos configurada en su cuenta BlazeSQL.\n* Documentación de la API de consulta en lenguaje natural de BlazeSQL: \n\n## Configuración\n\n1. **Clonar el repositorio:**\n\n ```bash\n git clone \n cd blaze-sql-mcp-server\n ```\n\n2. **Dependencias de instalación:**\n\n ```bash\n yarn install\n ```\n\n Esto instalará todas las dependencias necesarias, incluidas `@modelcontextprotocol/sdk` , `dotenv` y `zod` .\n\n3. **Configurar variables de entorno:**\n\n * Copie el archivo de entorno de ejemplo:\n\n ```bash\n cp .env.sample .env\n ```\n\n * Edite el archivo `.env` :\n\n ```dotenv\n # .env\n BLAZE_API_KEY=YOUR_BLAZESQL_API_KEY_HERE\n ```\n\n Reemplace `YOUR_BLAZESQL_API_KEY_HERE` con su clave API real obtenida de la configuración de su cuenta BlazeSQL.\n\n## Ejecución del servidor\n\n1. **Construir el servidor:** compilar el código TypeScript en JavaScript:\n\n ```bash\n yarn build\n ```\n\n2. **Ejecutar el servidor:** Ejecute el código compilado:\n\n ```bash\n node build/index.js\n ```\n\n El servidor se iniciará y registrará los mensajes en `stderr` (podría ver \"Clave API cargada correctamente...\", etc.). Ahora está a la espera de una conexión de cliente MCP mediante la entrada/salida estándar (stdio).\n\n## Conexión de un cliente MCP\n\nEste servidor utiliza el mecanismo de transporte **stdio** .\n\n### Uso del inspector MCP (recomendado para pruebas)\n\n1. Asegúrese de que el servidor **no** esté ejecutándose por separado.\n\n2. Ejecute el Inspector y dígale que inicie su servidor:\n\n ```bash\n npx @modelcontextprotocol/inspector node build/index.js\n ```\n\n3. Se iniciará la interfaz de usuario del Inspector y se conectará automáticamente a su servidor.\n\n4. Vaya a la pestaña \"Herramientas\" para interactuar con la herramienta `blazesql_query` .\n\n### Uso de clientes integrados (Cursor, Claude 3, etc.)\n\n1. **Inicie el servidor** en una terminal:\n\n ```bash\n node build/index.js\n ```\n\n2. **Configurar el cliente:** en la configuración de su cliente MCP, debe agregar una configuración de servidor personalizada.\n\n * **Transporte:** Seleccionar `stdio` .\n * **Comando:** Especifique el comando exacto para ejecutar el servidor. Debe proporcionar la **ruta absoluta** al nodo y la **ruta absoluta** al archivo `build/index.js` .\n * Ejemplo (macOS/Linux - ajuste las rutas según sea necesario): `/usr/local/bin/node /Users/your_username/path/to/blaze-sql-mcp-server/build/index.js`\n * Puedes encontrar la ruta al nodo usando `which node` en tu terminal.\n * Puede encontrar la ruta al proyecto usando `pwd` dentro del directorio del proyecto.\n * Guardar la configuración.\n\n3. El cliente ahora debería poder conectarse a su servidor local y enumerar/utilizar sus herramientas.\n\n## Uso de la herramienta `blazesql_query`\n\nUna vez conectado, el cliente puede llamar a la herramienta `blazesql_query` .\n\n* **Nombre de la herramienta:** `blazesql_query`\n\n* **Argumentos:**\n\n * `db_id` (cadena, obligatoria): El ID de la conexión a la base de datos de destino en su cuenta de BlazeSQL. Puede encontrar este ID en la aplicación web de BlazeSQL al administrar sus conexiones a la base de datos.\n * `natural_language_request` (string, required): La consulta que desea ejecutar, escrita en inglés simple (p. ej., \"muéstrame el número total de usuarios\"). *(La entrada se valida con `zod` ).*\n\n* **Ejemplo de llamada (utilizando la sintaxis `mcp test` para ilustración):**\n\n ```bash\n call-tool blazesql_query --db_id \"db_your_actual_db_id\" --natural_language_request \"What were the total sales last month?\"\n ```\n\n* **Salida:** si tiene éxito, la herramienta devuelve un solo bloque de contenido `text` que contiene:\n\n * La respuesta en lenguaje natural del agente BlazeSQL.\n * La consulta SQL generada dentro de una valla de código Markdown ( `sql ...` ).\n * Los datos resultan formateados como JSON dentro de un límite de código Markdown ( `json ...` ).\n\n Ejemplo de estructura dentro del bloque `text` :\n\n ````markdown\n **Agent Response:**\n The total sales last month were $12345.67.\n\n **Generated SQL:**\n ```sql\n SELECT sum(sales_amount) FROM sales WHERE sale_date >= date('now', '-1 month');\n ````\n\n **Resultado de los datos (JSON):**\n\n ```json\n [\n {\n \"sum(sales_amount)\": 12345.67\n }\n ]\n ```\n\n ```\n\n If unsuccessful, it returns a `text` content block containing the error message from the BlazeSQL API and marks the response as an error (`isError: true`).\n ```","Spanish","es-ES",{"_34":60,"_38":64,"_43":65},{"_36":61,"_38":62,"_40":63},"AI クライアントが BlazeSQL の自然言語クエリ API と対話し、SQL データベースに対する自然言語クエリを実行できるようにするモデルコンテキストプロトコルサーバー。","BlazeSQL MCP サーバー","# BlazeSQL MCP サーバー\n\nこのプロジェクトは、 `@modelcontextprotocol/sdk`を使用して、BlazeSQL 自然言語クエリ API のプロキシとして機能するモデルコンテキストプロトコル (MCP) サーバーを実装します。これにより、MCP 対応クライアント (Cursor、ツール使用機能を備えた Claude 3、MCP Inspector など) が自然言語を使用して BlazeSQL と対話できるようになります。\n\n## 特徴\n\n* MCP SDK の最新の`McpServer`ヘルパークラスを使用して構築されました。\n* BlazeSQL 自然言語クエリ API を`blazesql_query`という名前の MCP ツールとして公開します。\n* ツール入力パラメータの堅牢な検証に`zod`を使用します。\n* 環境変数を介して API キー認証を安全に処理します。\n* 標準の MCP stdio トランスポートを使用してクライアントと通信します。\n\n## ワークフロー図\n\nこの図は、クライアントが`blazesql_query`ツールを使用する場合の相互作用のシーケンスを示しています (注: 内部サーバーロジックでは`McpServer`が使用されるようになり、図に示されている低レベルのハンドラーに比べてツールの登録が簡素化されています)。\n\n```mermaid\nsequenceDiagram\n participant Client as MCP Client (e.g., Cursor)\n participant Server as BlazeSQL MCP Server (index.ts)\n participant Env as Environment (.env)\n participant BlazeAPI as BlazeSQL API\n\n Client->>Server: ListTools Request (via stdio)\n Server-->>Client: ListTools Response (tools: [blazesql_query]) (via stdio)\n\n Client->>Server: CallTool Request (blazesql_query, db_id, nl_request) (via stdio)\n Server->>Env: Read BLAZE_API_KEY\n Env-->>Server: BLAZE_API_KEY\n Server->>BlazeAPI: POST /natural_language_query_api (apiKey, db_id, nl_request)\n BlazeAPI->>BlazeAPI: Process Query (NL->SQL, Execute)\n BlazeAPI-->>Server: HTTPS Response (JSON: agent_response, query, data_result OR error)\n Server->>Server: Format Response (Agent response, SQL, and data into single text block)\n Server-->>Client: CallTool Response (content: [{type: text, text: formattedMarkdown}]) (via stdio)\n\n```\n\n## 前提条件\n\n* [Node.js](https://nodejs.org/) (LTS バージョンを推奨)\n* [糸](https://yarnpkg.com/)（クラシックまたはベリー）\n* API キーを持つ BlazeSQL アカウント (API には Team Advanced サブスクリプションが必要です)。\n* BlazeSQL アカウントに少なくとも 1 つのデータベース接続が構成されています。\n* BlazeSQL 自然言語クエリ API ドキュメント: \n\n## 設定\n\n1. **リポジトリをクローンします。**\n\n ```bash\n git clone \n cd blaze-sql-mcp-server\n ```\n\n2. **依存関係をインストールします:**\n\n ```bash\n yarn install\n ```\n\n これにより、 `@modelcontextprotocol/sdk` 、 `dotenv` 、 `zod`などの必要な依存関係がすべてインストールされます。\n\n3. **環境変数を設定します。**\n\n * サンプル環境ファイルをコピーします。\n\n ```bash\n cp .env.sample .env\n ```\n\n * `.env`ファイルを編集します。\n\n ```dotenv\n # .env\n BLAZE_API_KEY=YOUR_BLAZESQL_API_KEY_HERE\n ```\n\n `YOUR_BLAZESQL_API_KEY_HERE` 、BlazeSQL アカウント設定から取得した実際の API キーに置き換えます。\n\n## サーバーの実行\n\n1. **サーバーの構築:** TypeScript コードを JavaScript にコンパイルします。\n\n ```bash\n yarn build\n ```\n\n2. \\*\\*サーバーの実行:\\*\\*コンパイルされたコードを実行します。\n\n ```bash\n node build/index.js\n ```\n\n サーバーが起動し、 `stderr`にメッセージが記録されます（「APIキーが正常にロードされました...」などが表示される場合があります）。サーバーは標準入出力（stdio）経由でMCPクライアント接続を待機しています。\n\n## MCPクライアントの接続\n\nこのサーバーは**stdio トランス**ポートメカニズムを使用します。\n\n### MCP Inspector の使用 (テストに推奨)\n\n1. サーバーがすでに個別に実行されて**いない**ことを確認してください。\n\n2. Inspector を実行し、サーバーを起動するように指示します。\n\n ```bash\n npx @modelcontextprotocol/inspector node build/index.js\n ```\n\n3. Inspector UI が起動し、自動的にサーバーに接続します。\n\n4. `blazesql_query`ツールを操作するには、「ツール」タブに移動します。\n\n### 統合クライアントの使用 (Cursor、Claude 3 など)\n\n1. ターミナルで**サーバーを起動します**。\n\n ```bash\n node build/index.js\n ```\n\n2. **クライアントを構成する:** MCP クライアントの設定で、カスタムサーバー構成を追加する必要があります。\n\n * **トランスポート:** `stdio`を選択します。\n * **コマンド:**サーバーの実行に使用するコマンドを正確に指定します。node への**絶対パス**と`build/index.js`ファイルへの**絶対パス**を指定する必要があります。\n * 例 (macOS/Linux - 必要に応じてパスを調整してください): `/usr/local/bin/node /Users/your_username/path/to/blaze-sql-mcp-server/build/index.js`\n * ターミナルで`which node`を使用してノードへのパスを見つけることができます。\n * プロジェクトへのパスは、プロジェクトディレクトリ内で`pwd`を使用して見つけることができます。\n * 設定を保存します。\n\n3. これで、クライアントはローカルで実行されているサーバーに接続し、そのツールを一覧表示したり使用したりできるようになります。\n\n## `blazesql_query`ツールの使用\n\n接続されると、クライアントは`blazesql_query`ツールを呼び出すことができます。\n\n* **ツール名:** `blazesql_query`\n\n* **引数:**\n\n * `db_id` (文字列、必須): BlazeSQLアカウント内のターゲットデータベース接続のID。このIDは、データベース接続を管理する際にBlazeSQL Webアプリケーションで確認できます。\n * `natural_language_request` （文字列、必須）: 実行したいクエリを平易な英語で記述します（例：「ユーザーの合計数を表示してください」）。 *（入力は`zod`を使用して検証されます）*\n\n* **呼び出し例 (説明のために`mcp test`構文を使用):**\n\n ```bash\n call-tool blazesql_query --db_id \"db_your_actual_db_id\" --natural_language_request \"What were the total sales last month?\"\n ```\n\n* \\*\\*出力:\\*\\*成功した場合、ツールは次の内容を含む単一の`text`コンテンツブロックを返します。\n\n * BlazeSQL エージェントからの自然言語応答。\n * Markdown コードフェンス内で生成された SQL クエリ ( `sql ...` )。\n * データは、Markdown コードフェンス ( `json ...` ) 内で JSON としてフォーマットされます。\n\n `text`ブロック内の構造の例:\n\n ````markdown\n **Agent Response:**\n The total sales last month were $12345.67.\n\n **Generated SQL:**\n ```sql\n SELECT sum(sales_amount) FROM sales WHERE sale_date >= date('now', '-1 month');\n ````\n\n **データ結果 (JSON):**\n\n ```json\n [\n {\n \"sum(sales_amount)\": 12345.67\n }\n ]\n ```\n\n ```\n\n If unsuccessful, it returns a `text` content block containing the error message from the BlazeSQL API and marks the response as an error (`isError: true`).\n ```","Japanese","ja-JP",{"_34":67,"_38":71,"_43":72},{"_36":68,"_38":69,"_40":70},"AI 클라이언트가 BlazeSQL의 자연어 쿼리 API와 상호 작용할 수 있도록 하는 모델 컨텍스트 프로토콜 서버로, SQL 데이터베이스에 대한 자연어 쿼리를 허용합니다.","BlazeSQL MCP 서버","# BlazeSQL MCP 서버\n\n이 프로젝트는 `@modelcontextprotocol/sdk` 사용하여 BlazeSQL 자연어 쿼리 API의 프록시 역할을 하는 모델 컨텍스트 프로토콜(MCP) 서버를 구현합니다. MCP 호환 클라이언트(예: Cursor, 도구를 사용하는 Claude 3, MCP Inspector 등)가 자연어를 사용하여 BlazeSQL과 상호 작용할 수 있도록 지원합니다.\n\n## 특징\n\n* MCP SDK의 최신 `McpServer` 도우미 클래스를 사용하여 빌드되었습니다.\n* BlazeSQL 자연어 쿼리 API를 `blazesql_query` 라는 이름의 MCP 도구로 공개합니다.\n* `zod` 사용하여 도구 입력 매개변수의 견고한 검증을 실시합니다.\n* 환경 변수를 통해 API 키 인증을 안전하게 처리합니다.\n* 표준 MCP stdio 전송을 사용하여 클라이언트와 통신합니다.\n\n## 워크플로 다이어그램\n\n이 다이어그램은 클라이언트가 `blazesql_query` 도구를 사용할 때의 상호작용 순서를 보여줍니다(참고: 내부 서버 로직은 이제 `McpServer` 사용하는데, 이는 다이어그램에 표시된 저수준 핸들러에 비해 도구 등록을 간소화합니다).\n\n지엑스피1\n\n## 필수 조건\n\n* [Node.js](https://nodejs.org/) (LTS 버전 권장)\n* [원사](https://yarnpkg.com/) (클래식 또는 베리)\n* API 키가 있는 BlazeSQL 계정(API를 사용하려면 Team Advanced 구독이 필요함).\n* BlazeSQL 계정에 하나 이상의 데이터베이스 연결이 구성되어 있습니다.\n* BlazeSQL 자연어 쿼리 API 문서: \n\n## 설정\n\n1. **저장소 복제:**\n\n ```bash\n git clone \n cd blaze-sql-mcp-server\n ```\n\n2. **종속성 설치:**\n\n ```bash\n yarn install\n ```\n\n 이렇게 하면 `@modelcontextprotocol/sdk` , `dotenv` , `zod` 포함한 모든 필수 종속성이 설치됩니다.\n\n3. **환경 변수 구성:**\n\n * 예제 환경 파일을 복사하세요.\n\n ```bash\n cp .env.sample .env\n ```\n\n * `.env` 파일을 편집합니다.\n\n ```dotenv\n # .env\n BLAZE_API_KEY=YOUR_BLAZESQL_API_KEY_HERE\n ```\n\n `YOUR_BLAZESQL_API_KEY_HERE` BlazeSQL 계정 설정에서 얻은 실제 API 키로 바꾸세요.\n\n## 서버 실행\n\n1. **서버 빌드:** TypeScript 코드를 JavaScript로 컴파일합니다.\n\n ```bash\n yarn build\n ```\n\n2. **서버 실행:** 컴파일된 코드를 실행합니다.\n\n ```bash\n node build/index.js\n ```\n\n 서버가 시작되고 `stderr` 에 메시지가 기록됩니다(\"API 키가 성공적으로 로드되었습니다...\" 등의 메시지가 표시될 수 있음). 이제 표준 입출력(stdio)을 통해 MCP 클라이언트 연결을 수신합니다.\n\n## MCP 클라이언트 연결\n\n이 서버는 **stdio** 전송 메커니즘을 사용합니다.\n\n### MCP Inspector 사용(테스트 권장)\n\n1. 서버가 별도로 실행되고 있지 **않은지** 확인하세요.\n\n2. 검사기를 실행하여 서버를 시작하도록 지시합니다.\n\n ```bash\n npx @modelcontextprotocol/inspector node build/index.js\n ```\n\n3. Inspector UI가 실행되고 자동으로 서버에 연결됩니다.\n\n4. \"도구\" 탭으로 이동하여 `blazesql_query` 도구와 상호 작용하세요.\n\n### 통합 클라이언트(Cursor, Claude 3 등) 사용\n\n1. 터미널에서 **서버를 시작합니다** .\n\n ```bash\n node build/index.js\n ```\n\n2. **클라이언트 구성:** MCP 클라이언트 설정에서 사용자 지정 서버 구성을 추가해야 합니다.\n\n * **전송:** `stdio` 선택하세요.\n * **명령어:** 서버 실행에 사용되는 정확한 명령어를 지정하세요. node의 **절대 경로** 와 `build/index.js` 파일의 **절대 경로를** 제공해야 합니다.\n * 예시(macOS/Linux - 필요에 따라 경로 조정): `/usr/local/bin/node /Users/your_username/path/to/blaze-sql-mcp-server/build/index.js`\n * 터미널에서 `which node` 입력하면 노드 경로를 찾을 수 있습니다.\n * 프로젝트 디렉토리 내에서 `pwd` 사용하여 프로젝트 경로를 찾을 수 있습니다.\n * 구성을 저장합니다.\n\n3. 이제 클라이언트는 로컬에서 실행 중인 서버에 연결하여 해당 도구를 나열하고 사용할 수 있습니다.\n\n## `blazesql_query` 도구 사용\n\n연결되면 클라이언트는 `blazesql_query` 도구를 호출할 수 있습니다.\n\n* **도구 이름:** `blazesql_query`\n\n* **인수:**\n\n * `db_id` (문자열, 필수): BlazeSQL 계정의 대상 데이터베이스 연결 ID입니다. BlazeSQL 웹 애플리케이션에서 데이터베이스 연결을 관리할 때 이 ID를 찾을 수 있습니다.\n * `natural_language_request` (문자열, 필수): 실행하려는 쿼리를 일반 영어로 작성합니다(예: \"사용자 총 수를 보여주세요\"). *(입력은 `zod` 사용하여 검증됩니다)*\n\n* **예제 호출(설명을 위해 `mcp test` 구문 사용):**\n\n ```bash\n call-tool blazesql_query --db_id \"db_your_actual_db_id\" --natural_language_request \"What were the total sales last month?\"\n ```\n\n* **출력:** 성공하면 도구는 다음을 포함하는 단일 `text` 콘텐츠 블록을 반환합니다.\n\n * BlazeSQL 에이전트의 자연어 응답입니다.\n * Markdown 코드 펜스( `sql ...` ) 내에서 생성된 SQL 쿼리입니다.\n * 데이터 결과는 Markdown 코드 펜스( `json ...` ) 내에서 JSON으로 포맷됩니다.\n\n `text` 블록 내의 구조 예:\n\n ````markdown\n **Agent Response:**\n The total sales last month were $12345.67.\n\n **Generated SQL:**\n ```sql\n SELECT sum(sales_amount) FROM sales WHERE sale_date >= date('now', '-1 month');\n ````\n\n **데이터 결과(JSON):**\n\n ```json\n [\n {\n \"sum(sales_amount)\": 12345.67\n }\n ]\n ```\n\n ```\n\n If unsuccessful, it returns a `text` content block containing the error message from the BlazeSQL API and marks the response as an error (`isError: true`).\n ```","Korean","ko-KR",{"_34":74,"_38":78,"_43":79},{"_36":75,"_38":76,"_40":77},"Сервер Model Context Protocol, который позволяет клиентам ИИ взаимодействовать с API запросов на естественном языке BlazeSQL, позволяя выполнять запросы на естественном языке к базам данных SQL.","MCP-сервер BlazeSQL","# MCP-сервер BlazeSQL\n\nЭтот проект реализует сервер Model Context Protocol (MCP) с использованием `@modelcontextprotocol/sdk` , который действует как прокси для API BlazeSQL Natural Language Query. Он позволяет MCP-совместимым клиентам (например, Cursor, Claude 3 с использованием инструмента, MCP Inspector и т. д.) взаимодействовать с BlazeSQL с использованием естественного языка.\n\n## Функции\n\n* Создан с использованием современного вспомогательного класса `McpServer` из MCP SDK.\n* Предоставляет API запросов на естественном языке BlazeSQL как инструмент MCP с именем `blazesql_query` .\n* Использует `zod` для надежной проверки входных параметров инструмента.\n* Обеспечивает безопасную аутентификацию API-ключа с помощью переменных среды.\n* Взаимодействует с клиентами, используя стандартный транспорт MCP stdio.\n\n## Диаграмма рабочего процесса\n\nНа этой диаграмме показана последовательность взаимодействий, когда клиент использует инструмент `blazesql_query` (Примечание: внутренняя логика сервера теперь использует `McpServer` , что упрощает регистрацию инструмента по сравнению с низкоуровневыми обработчиками, показанными на диаграмме):\n\n```mermaid\nsequenceDiagram\n participant Client as MCP Client (e.g., Cursor)\n participant Server as BlazeSQL MCP Server (index.ts)\n participant Env as Environment (.env)\n participant BlazeAPI as BlazeSQL API\n\n Client->>Server: ListTools Request (via stdio)\n Server-->>Client: ListTools Response (tools: [blazesql_query]) (via stdio)\n\n Client->>Server: CallTool Request (blazesql_query, db_id, nl_request) (via stdio)\n Server->>Env: Read BLAZE_API_KEY\n Env-->>Server: BLAZE_API_KEY\n Server->>BlazeAPI: POST /natural_language_query_api (apiKey, db_id, nl_request)\n BlazeAPI->>BlazeAPI: Process Query (NL->SQL, Execute)\n BlazeAPI-->>Server: HTTPS Response (JSON: agent_response, query, data_result OR error)\n Server->>Server: Format Response (Agent response, SQL, and data into single text block)\n Server-->>Client: CallTool Response (content: [{type: text, text: formattedMarkdown}]) (via stdio)\n\n```\n\n## Предпосылки\n\n* [Node.js](https://nodejs.org/) (рекомендуется версия LTS)\n* [Пряжа](https://yarnpkg.com/) (классическая или ягодная)\n* Учетная запись BlazeSQL с ключом API (для API требуется подписка Team Advanced).\n* В вашей учетной записи BlazeSQL настроено как минимум одно подключение к базе данных.\n* Документация API запросов на естественном языке BlazeSQL: \n\n## Настраивать\n\n1. **Клонировать репозиторий:**\n\n ```bash\n git clone \n cd blaze-sql-mcp-server\n ```\n\n2. **Установить зависимости:**\n\n ```bash\n yarn install\n ```\n\n Это установит все необходимые зависимости, включая `@modelcontextprotocol/sdk` , `dotenv` и `zod` .\n\n3. **Настройте переменные среды:**\n\n * Скопируйте пример файла среды:\n\n ```bash\n cp .env.sample .env\n ```\n\n * Отредактируйте файл `.env` :\n\n ```dotenv\n # .env\n BLAZE_API_KEY=YOUR_BLAZESQL_API_KEY_HERE\n ```\n\n Замените `YOUR_BLAZESQL_API_KEY_HERE` на ваш фактический ключ API, полученный из настроек вашей учетной записи BlazeSQL.\n\n## Запуск сервера\n\n1. **Сборка сервера:** Скомпилируйте код TypeScript в JavaScript:\n\n ```bash\n yarn build\n ```\n\n2. **Запустите сервер:** Выполните скомпилированный код:\n\n ```bash\n node build/index.js\n ```\n\n Сервер запустится и запишет сообщения в `stderr` (вы можете увидеть «API Key загружен успешно...» и т. д.). Теперь он прослушивает клиентское соединение MCP через стандартный ввод/вывод (stdio).\n\n## Подключение MCP-клиента\n\nЭтот сервер использует механизм транспорта **stdio** .\n\n### Использование MCP Inspector (рекомендуется для тестирования)\n\n1. Убедитесь, что сервер еще **не** запущен отдельно.\n\n2. Запустите Inspector и дайте ему команду запустить ваш сервер:\n\n ```bash\n npx @modelcontextprotocol/inspector node build/index.js\n ```\n\n3. Запустится пользовательский интерфейс Inspector, который автоматически подключится к вашему серверу.\n\n4. Перейдите на вкладку «Инструменты», чтобы взаимодействовать с инструментом `blazesql_query` .\n\n### Использование интегрированных клиентов (Cursor, Claude 3 и т. д.)\n\n1. **Запустите сервер** в терминале:\n\n ```bash\n node build/index.js\n ```\n\n2. **Настройте клиент:** в настройках вашего клиента MCP вам необходимо добавить пользовательскую конфигурацию сервера.\n\n * **Транспорт:** Выберите `stdio` .\n * **Команда:** Укажите точную команду, используемую для запуска сервера. Вам необходимо указать **абсолютный путь** к узлу и **абсолютный путь** к файлу `build/index.js` .\n * Пример (macOS/Linux — при необходимости измените пути): `/usr/local/bin/node /Users/your_username/path/to/blaze-sql-mcp-server/build/index.js`\n * Путь к узлу можно найти, используя `which node` в вашем терминале.\n * Путь к проекту можно найти с помощью `pwd` внутри каталога проекта.\n * Сохраните конфигурацию.\n\n3. Теперь клиент сможет подключиться к локально работающему серверу и просмотреть/использовать его инструменты.\n\n## Использование инструмента `blazesql_query`\n\nПосле подключения клиент может вызвать инструмент `blazesql_query` .\n\n* **Имя инструмента:** `blazesql_query`\n\n* **Аргументы:**\n\n * `db_id` (string, required): Идентификатор целевого подключения к базе данных в вашей учетной записи BlazeSQL. Вы можете найти этот идентификатор в веб-приложении BlazeSQL при управлении подключениями к базе данных.\n * `natural_language_request` (string, required): Запрос, который вы хотите выполнить, написанный на простом английском языке (например, «покажите мне общее количество пользователей»). *(Входные данные проверяются с помощью `zod` )*\n\n* **Пример вызова (с использованием синтаксиса `mcp test` для иллюстрации):**\n\n ```bash\n call-tool blazesql_query --db_id \"db_your_actual_db_id\" --natural_language_request \"What were the total sales last month?\"\n ```\n\n* **Вывод:** в случае успеха инструмент возвращает один блок `text` содержимого, содержащий:\n\n * Ответ на естественном языке от агента BlazeSQL.\n * Сгенерированный SQL-запрос внутри кода Markdown ( `sql ...` ).\n * Результаты данных форматируются как JSON в коде Markdown ( `json ...` ).\n\n Пример структуры внутри `text` блока:\n\n ````markdown\n **Agent Response:**\n The total sales last month were $12345.67.\n\n **Generated SQL:**\n ```sql\n SELECT sum(sales_amount) FROM sales WHERE sale_date >= date('now', '-1 month');\n ````\n\n **Результат данных (JSON):**\n\n ```json\n [\n {\n \"sum(sales_amount)\": 12345.67\n }\n ]\n ```\n\n ```\n\n If unsuccessful, it returns a `text` content block containing the error message from the BlazeSQL API and marks the response as an error (`isError: true`).\n ```","Russian","ru-RU",{"_34":81,"_38":85,"_43":86},{"_36":82,"_38":83,"_40":84},"模型上下文协议服务器使 AI 客户端能够与 BlazeSQL 的自然语言查询 API 进行交互，从而允许对 SQL 数据库进行自然语言查询。","BlazeSQL MCP 服务器","# BlazeSQL MCP 服务器\n\n该项目使用`@modelcontextprotocol/sdk`实现了一个模型上下文协议 (MCP) 服务器，该服务器充当 BlazeSQL 自然语言查询 API 的代理。它允许兼容 MCP 的客户端（例如 Cursor、使用工具的 Claude 3、MCP Inspector 等）使用自然语言与 BlazeSQL 进行交互。\n\n## 特征\n\n* 使用 MCP SDK 中的现代`McpServer`帮助类构建。\n* 将 BlazeSQL 自然语言查询 API 作为名为`blazesql_query`的 MCP 工具公开。\n* 使用`zod`对工具输入参数进行稳健验证。\n* 通过环境变量安全地处理 API 密钥认证。\n* 使用标准 MCP stdio 传输与客户端通信。\n\n## 工作流程图\n\n该图显示了客户端使用`blazesql_query`工具时的交互顺序（注意：内部服务器逻辑现在使用`McpServer` ，与图中所示的低级处理程序相比，它简化了工具注册）：\n\n```mermaid\nsequenceDiagram\n participant Client as MCP Client (e.g., Cursor)\n participant Server as BlazeSQL MCP Server (index.ts)\n participant Env as Environment (.env)\n participant BlazeAPI as BlazeSQL API\n\n Client->>Server: ListTools Request (via stdio)\n Server-->>Client: ListTools Response (tools: [blazesql_query]) (via stdio)\n\n Client->>Server: CallTool Request (blazesql_query, db_id, nl_request) (via stdio)\n Server->>Env: Read BLAZE_API_KEY\n Env-->>Server: BLAZE_API_KEY\n Server->>BlazeAPI: POST /natural_language_query_api (apiKey, db_id, nl_request)\n BlazeAPI->>BlazeAPI: Process Query (NL->SQL, Execute)\n BlazeAPI-->>Server: HTTPS Response (JSON: agent_response, query, data_result OR error)\n Server->>Server: Format Response (Agent response, SQL, and data into single text block)\n Server-->>Client: CallTool Response (content: [{type: text, text: formattedMarkdown}]) (via stdio)\n\n```\n\n## 先决条件\n\n* [Node.js](https://nodejs.org/) （推荐 LTS 版本）\n* [纱线](https://yarnpkg.com/)（经典或浆果）\n* 具有 API 密钥的 BlazeSQL 帐户（API 需要 Team Advanced 订阅）。\n* 您的 BlazeSQL 帐户中至少配置了一个数据库连接。\n* BlazeSQL 自然语言查询 API 文档： \n\n## 设置\n\n1. **克隆存储库：**\n\n ```bash\n git clone \n cd blaze-sql-mcp-server\n ```\n\n2. **安装依赖项：**\n\n ```bash\n yarn install\n ```\n\n 这将安装所有必要的依赖项，包括`@modelcontextprotocol/sdk` 、 `dotenv`和`zod` 。\n\n3. **配置环境变量：**\n\n * 复制示例环境文件：\n\n ```bash\n cp .env.sample .env\n ```\n\n * 编辑`.env`文件：\n\n ```dotenv\n # .env\n BLAZE_API_KEY=YOUR_BLAZESQL_API_KEY_HERE\n ```\n\n 将`YOUR_BLAZESQL_API_KEY_HERE`替换为从 BlazeSQL 帐户设置中获取的实际 API 密钥。\n\n## 运行服务器\n\n1. \\*\\*构建服务器：\\*\\*将 TypeScript 代码编译为 JavaScript：\n\n ```bash\n yarn build\n ```\n\n2. \\*\\*运行服务器：\\*\\*执行编译后的代码：\n\n ```bash\n node build/index.js\n ```\n\n 服务器将启动并将消息记录到`stderr` （您可能会看到“API 密钥已成功加载……”等信息）。现在，服务器正在通过标准输入/输出 (stdio) 监听 MCP 客户端连接。\n\n## 连接 MCP 客户端\n\n该服务器使用**stdio**传输机制。\n\n### 使用 MCP 检查器（推荐用于测试）\n\n1. 确保服务器**尚未**单独运行。\n\n2. 运行检查器，告诉它启动你的服务器：\n\n ```bash\n npx @modelcontextprotocol/inspector node build/index.js\n ```\n\n3. Inspector UI 将启动并自动连接到您的服务器。\n\n4. 导航到“工具”选项卡以与`blazesql_query`工具进行交互。\n\n### 使用集成客户端（Cursor、Claude 3 等）\n\n1. 在终端中**启动服务器**：\n\n ```bash\n node build/index.js\n ```\n\n2. \\*\\*配置客户端：\\*\\*在您的 MCP 客户端设置中，您需要添加自定义服务器配置。\n\n * \\*\\*传输：\\*\\*选择`stdio` 。\n * **命令：**指定用于运行服务器的具体命令。您需要提供 node 的**绝对路径**以及`build/index.js`文件的**绝对路径**。\n * 示例（macOS/Linux - 根据需要调整路径）： `/usr/local/bin/node /Users/your_username/path/to/blaze-sql-mcp-server/build/index.js`\n * 您可以使用终端中的`which node`找到节点的路径。\n * 您可以在项目目录中使用`pwd`找到项目的路径。\n * 保存配置。\n\n3. 客户端现在应该能够连接到您本地运行的服务器并列出/使用其工具。\n\n## 使用`blazesql_query`工具\n\n一旦连接，客户端就可以调用`blazesql_query`工具。\n\n* **工具名称：** `blazesql_query`\n\n* **参数：**\n\n * `db_id` （字符串，必需）：BlazeSQL 帐户中目标数据库连接的 ID。您可以在 BlazeSQL Web 应用程序管理数据库连接时找到此 ID。\n * `natural_language_request` （字符串，必需）：您要执行的查询，以简单的英语编写（例如，“显示用户总数”）。 *（使用`zod`验证输入）*\n\n* **示例调用（使用`mcp test`语法进行说明）：**\n\n ```bash\n call-tool blazesql_query --db_id \"db_your_actual_db_id\" --natural_language_request \"What were the total sales last month?\"\n ```\n\n* \\*\\*输出：\\*\\*如果成功，该工具将返回包含以下内容的单个`text`内容块：\n\n * BlazeSQL 代理的自然语言响应。\n * Markdown 代码围栏内生成的 SQL 查询 ( `sql ...` )。\n * 数据结果在 Markdown 代码围栏 ( `json ...` ) 内格式化为 JSON。\n\n `text`块内的示例结构：\n\n ````markdown\n **Agent Response:**\n The total sales last month were $12345.67.\n\n **Generated SQL:**\n ```sql\n SELECT sum(sales_amount) FROM sales WHERE sale_date >= date('now', '-1 month');\n ````\n\n **数据结果（JSON）：**\n\n ```json\n [\n {\n \"sum(sales_amount)\": 12345.67\n }\n ]\n ```\n\n ```\n\n If unsuccessful, it returns a `text` content block containing the error message from the BlazeSQL API and marks the response as an error (`isError: true`).\n ```","Chinese","zh-CN","githubUser","locale",{"_90":91,"_20":92,"_101":47,"_102":47,"_23":103,"_104":105,"_38":48,"_130":131,"_138":-5,"_139":-5,"_140":141,"_184":185,"_190":-5,"_112":168,"_191":192,"_8":193,"_194":195},"addedAt","2025-05-05T02:00:46.917933Z",[93,94,95,96,97,98,99,100],"category:databases","category:search","environment:network","hosting:remote-capable","language:javascript","security_grade:a","quality_grade:a","license_grade:f","descriptionMarkdown","descriptionPlainText",6291,"integrations",[106,116,123],{"_107":108,"_36":114,"_8":115},"brand",{"_109":110,"_38":111,"_112":113},"logoSvgPath","M24 0v24H0V0h24ZM10.933 15.89H6.84v5.52h4.198v-.93H7.955v-1.503h2.77v-.93h-2.77v-1.224h2.978v-.934Zm2.146 0h-1.084v5.52h1.035v-3.6l2.226 3.6h1.118v-5.52h-1.036v3.686l-2.259-3.687Zm5.117 0h-1.208l1.973 5.52h1.19l1.976-5.52h-1.182l-1.352 4.085-1.397-4.086ZM5.4 19.68H3.72v1.68H5.4v-1.68Z",".ENV","slug","dotenv","Handles API key authentication securely via environment variables for connecting to the BlazeSQL service.","muspu6c06f",{"_107":117,"_36":121,"_8":122},{"_109":118,"_38":119,"_112":120},"M11.998,24c-0.321,0-0.641-0.084-0.922-0.247l-2.936-1.737c-0.438-0.245-0.224-0.332-0.08-0.383 c0.585-0.203,0.703-0.25,1.328-0.604c0.065-0.037,0.151-0.023,0.218,0.017l2.256,1.339c0.082,0.045,0.197,0.045,0.272,0l8.795-5.076 c0.082-0.047,0.134-0.141,0.134-0.238V6.921c0-0.099-0.053-0.192-0.137-0.242l-8.791-5.072c-0.081-0.047-0.189-0.047-0.271,0 L3.075,6.68C2.99,6.729,2.936,6.825,2.936,6.921v10.15c0,0.097,0.054,0.189,0.139,0.235l2.409,1.392 c1.307,0.654,2.108-0.116,2.108-0.89V7.787c0-0.142,0.114-0.253,0.256-0.253h1.115c0.139,0,0.255,0.112,0.255,0.253v10.021 c0,1.745-0.95,2.745-2.604,2.745c-0.508,0-0.909,0-2.026-0.551L2.28,18.675c-0.57-0.329-0.922-0.945-0.922-1.604V6.921 c0-0.659,0.353-1.275,0.922-1.603l8.795-5.082c0.557-0.315,1.296-0.315,1.848,0l8.794,5.082c0.57,0.329,0.924,0.944,0.924,1.603 v10.15c0,0.659-0.354,1.273-0.924,1.604l-8.794,5.078C12.643,23.916,12.324,24,11.998,24z M19.099,13.993 c0-1.9-1.284-2.406-3.987-2.763c-2.731-0.361-3.009-0.548-3.009-1.187c0-0.528,0.235-1.233,2.258-1.233 c1.807,0,2.473,0.389,2.747,1.607c0.024,0.115,0.129,0.199,0.247,0.199h1.141c0.071,0,0.138-0.031,0.186-0.081 c0.048-0.054,0.074-0.123,0.067-0.196c-0.177-2.098-1.571-3.076-4.388-3.076c-2.508,0-4.004,1.058-4.004,2.833 c0,1.925,1.488,2.457,3.895,2.695c2.88,0.282,3.103,0.703,3.103,1.269c0,0.983-0.789,1.402-2.642,1.402 c-2.327,0-2.839-0.584-3.011-1.742c-0.02-0.124-0.126-0.215-0.253-0.215h-1.137c-0.141,0-0.254,0.112-0.254,0.253 c0,1.482,0.806,3.248,4.655,3.248C17.501,17.007,19.099,15.91,19.099,13.993z","Node.js","node-dot-js","Provides integration with BlazeSQL Natural Language Query API, allowing users to query databases using natural language requests and receive SQL queries and formatted results.","kb0rks0n71",{"_107":124,"_36":128,"_8":129},{"_109":125,"_38":126,"_112":127},"M1.125 0C.502 0 0 .502 0 1.125v21.75C0 23.498.502 24 1.125 24h21.75c.623 0 1.125-.502 1.125-1.125V1.125C24 .502 23.498 0 22.875 0zm17.363 9.75c.612 0 1.154.037 1.627.111a6.38 6.38 0 0 1 1.306.34v2.458a3.95 3.95 0 0 0-.643-.361 5.093 5.093 0 0 0-.717-.26 5.453 5.453 0 0 0-1.426-.2c-.3 0-.573.028-.819.086a2.1 2.1 0 0 0-.623.242c-.17.104-.3.229-.393.374a.888.888 0 0 0-.14.49c0 .196.053.373.156.529.104.156.252.304.443.444s.423.276.696.41c.273.135.582.274.926.416.47.197.892.407 1.266.628.374.222.695.473.963.753.268.279.472.598.614.957.142.359.214.776.214 1.253 0 .657-.125 1.21-.373 1.656a3.033 3.033 0 0 1-1.012 1.085 4.38 4.38 0 0 1-1.487.596c-.566.12-1.163.18-1.79.18a9.916 9.916 0 0 1-1.84-.164 5.544 5.544 0 0 1-1.512-.493v-2.63a5.033 5.033 0 0 0 3.237 1.2c.333 0 .624-.03.872-.09.249-.06.456-.144.623-.25.166-.108.29-.234.373-.38a1.023 1.023 0 0 0-.074-1.089 2.12 2.12 0 0 0-.537-.5 5.597 5.597 0 0 0-.807-.444 27.72 27.72 0 0 0-1.007-.436c-.918-.383-1.602-.852-2.053-1.405-.45-.553-.676-1.222-.676-2.005 0-.614.123-1.141.369-1.582.246-.441.58-.804 1.004-1.089a4.494 4.494 0 0 1 1.47-.629 7.536 7.536 0 0 1 1.77-.201zm-15.113.188h9.563v2.166H9.506v9.646H6.789v-9.646H3.375z","TypeScript","typescript","Used to implement the MCP server that connects to BlazeSQL's API, enabling natural language database querying capabilities.","wqhapgqa6a","namespace",{"_132":133,"_112":137},"owner",{"_134":135},"githubAccount",{"_136":137},"login","arjshiv","readmeMarkdown","redditPostExternalId","repository",{"_142":143,"_153":154,"_157":158,"_175":-5,"_176":177,"_178":-5,"_179":180},"argumentsJsonSchema",{"_144":145,"_150":151,"_12":152},"properties",{"_146":147},"BLAZE_API_KEY",{"_36":148,"_12":149},"Your BlazeSQL API key obtained from your BlazeSQL account settings","string","required",[146],"object","githubProject",{"_40":49,"_155":156},"url","https://github.com/arjshiv/blaze-sql-mcp-server","githubRepository",{"_159":160,"_161":162,"_163":164,"_23":165,"_166":167,"_38":168,"_132":169,"_171":172,"_173":-5,"_174":172},"defaultBranch","main","fullName","arjshiv/blaze-sql-mcp-server","githubTreeId",368798,12305,"language","JavaScript","blaze-sql-mcp-server",{"_136":137,"_8":170},"gtk987ta7b","recentStargazers",0,"spdxLicense","stargazers","glamaJson","inspectable",true,"npmPackage","supportedPlatforms",[181,182,183],"WINDOWS","LINUX","MACOS","scores",{"_186":172,"_187":188,"_189":188},"license","quality",100,"security","simpleIcon","toolCount",1,"y6icx0ab46","updatedAt","2025-10-02T15:16:05.675824Z","meta",[198,200,202,209,212,214,217,220,223,226,229,231,234,237,239,241,243,245,247],{"_10":199},"BlazeSQL MCP Server | Glama",{"_201":47,"_38":36},"content",{"_203":204,"_205":206,"_207":208},"href","https://glama.ai/mcp/servers/@arjshiv/blaze-sql-mcp-server","rel","canonical","tagName","link",{"_201":48,"_210":211},"property","og:title",{"_201":47,"_210":213},"og:description",{"_201":215,"_210":216},"https://glama.ai/logo.png","og:logo",{"_201":218,"_210":219},"https://glama.ai/mcp/servers/@arjshiv/blaze-sql-mcp-server/og-image","og:image",{"_201":221,"_210":222},"630","og:image:height",{"_201":224,"_210":225},"1200","og:image:width",{"_201":227,"_210":228},"website","og:type",{"_201":204,"_210":230},"og:url",{"_201":232,"_210":233},"Glama – MCP Hosting Platform","og:site_name",{"_201":235,"_38":236},"summary_large_image","twitter:card",{"_201":232,"_38":238},"twitter:title",{"_201":47,"_38":240},"twitter:description",{"_201":218,"_38":242},"twitter:image",{"_201":221,"_38":244},"twitter:image:height",{"_201":224,"_38":246},"twitter:image:width",{"_248":249},"script:ld+json",{"_250":251,"_252":253},"@context","http://schema.org","@graph",[254],{"_250":255,"_256":257,"_258":259},"https://schema.org","@type","BreadcrumbList","itemListElement",[260,266,270,274],{"_256":261,"_262":263,"_38":264,"_265":192},"ListItem","item","/","Glama","position",{"_256":261,"_262":267,"_38":268,"_265":269},"/mcp","MCP",2,{"_256":261,"_262":271,"_38":272,"_265":273},"/mcp/servers","Servers",3,{"_256":261,"_262":275,"_38":48,"_265":276},"/mcp/servers/@arjshiv/blaze-sql-mcp-server",4,"visitorIsMaintainer",false,"routes/_public/mcp/servers/~namespace/~slug/_pages/_explorer/blob/~commit/~/_route",{"_281":282,"_34":288,"_310":311,"_196":322,"_320":346,"_347":348},"blob",{"_201":283,"_284":285,"_286":287},"# Node\nnode_modules/\n\n# Logs\n*.log\n\n# Environment variables\n.env\n.env.*\n!.env.sample\n\n# Editor/Tool History\n.history/\n","fileName",".gitignore","size",118,{"_90":91,"_20":289,"_101":47,"_102":47,"_23":103,"_104":290,"_38":48,"_130":297,"_138":-5,"_139":-5,"_140":300,"_184":309,"_190":-5,"_112":168,"_191":192,"_8":193,"_194":195},[93,94,95,96,97,98,99,100],[291,293,295],{"_107":292,"_36":114,"_8":115},{"_109":110,"_38":111,"_112":113},{"_107":294,"_36":121,"_8":122},{"_109":118,"_38":119,"_112":120},{"_107":296,"_36":128,"_8":129},{"_109":125,"_38":126,"_112":127},{"_132":298,"_112":137},{"_134":299},{"_136":137},{"_142":301,"_153":305,"_157":306,"_175":-5,"_176":177,"_178":-5,"_179":308},{"_144":302,"_150":304,"_12":152},{"_146":303},{"_36":148,"_12":149},[146],{"_40":49,"_155":156},{"_159":160,"_161":162,"_163":164,"_23":165,"_166":167,"_38":168,"_132":307,"_171":172,"_173":-5,"_174":172},{"_136":137,"_8":170},[181,182,183],{"_186":172,"_187":188,"_189":188},"member",{"_312":313,"_316":317,"_318":285,"_163":164,"_319":278,"_320":285,"_321":177,"_286":287,"_12":281},"commit",{"_314":315},"sha","cfe077acfce151f4f399d5d6a8adcf2f13baa275","contentStoredObjectPointerId",10232951,"fullPath","indexable","path","safeToDisplayAsText",[323,325,327,329,331,332,333,335,336,337,338,339,340,341,342,343,344,345],{"_10":324},".gitignore | BlazeSQL MCP Server | Glama",{"_201":326,"_38":36},"Enterprise-grade security, privacy, and features like agents, MCP, prompt templates, and more.",{"_203":328,"_205":206,"_207":208},"https://glama.ai/mcp/servers/@arjshiv/blaze-sql-mcp-server/blob/cfe077acfce151f4f399d5d6a8adcf2f13baa275/",{"_201":330,"_210":211},".gitignore | BlazeSQL MCP Server",{"_201":326,"_210":213},{"_201":215,"_210":216},{"_201":334,"_210":219},"https://glama.ai/generated-images/og?title=.gitignore+%7C+BlazeSQL+MCP+Server",{"_201":221,"_210":222},{"_201":224,"_210":225},{"_201":227,"_210":228},{"_201":328,"_210":230},{"_201":232,"_210":233},{"_201":235,"_38":236},{"_201":232,"_38":238},{"_201":326,"_38":240},{"_201":334,"_38":242},{"_201":221,"_38":244},{"_201":224,"_38":246},"","treeMembers",[349,353,356,359,362,364,367,370,373],{"_312":350,"_318":351,"_320":351,"_12":352},{"_314":315},"build","tree",{"_312":354,"_318":355,"_320":355,"_12":352},{"_314":315},".cursor",{"_312":357,"_318":358,"_320":358,"_12":352},{"_314":315},"src",{"_312":360,"_318":361,"_320":361,"_12":281},{"_314":315},".env.sample",{"_312":363,"_318":285,"_320":285,"_12":281},{"_314":315},{"_312":365,"_318":366,"_320":366,"_12":281},{"_314":315},"package.json",{"_312":368,"_318":369,"_320":369,"_12":281},{"_314":315},"README.md",{"_312":371,"_318":372,"_320":372,"_12":281},{"_314":315},"tsconfig.json",{"_312":374,"_318":375,"_320":375,"_12":281},{"_314":315},"yarn.lock","actionData","errors"]

MCP directory API