[{"_1":2,"_504":-5,"_505":-5},"loaderData",{"_3":4,"_22":23,"_24":25,"_376":377},"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",6153842,"membership","referenceId","0196b37d-1b65-705f-a96d-9e9a41c354e8","userAccount","routes/_public/_layout",{},"routes/_public/mcp/servers/~namespace/~slug/_pages/_layout",{"_26":27,"_68":-5,"_69":28,"_29":70,"_294":295,"_375":212},"availableLocales",[28,40,47,54,61],{"_29":30,"_33":37,"_38":39},"mcpServer",{"_31":32,"_33":34,"_35":36},"description","An API server that implements the Model Completion Protocol (MCP) for Cryo blockchain data extraction, allowing users to query Ethereum blockchain data through any MCP-compatible client.","name","Cryo MCP Server","readme","# Cryo MCP 🧊\n\nA Model Completion Protocol (MCP) server for the [Cryo](https://github.com/paradigmxyz/cryo) blockchain data extraction tool. \n\nCryo MCP allows you to access Cryo's powerful blockchain data extraction capabilities via an API server that implements the MCP protocol, making it easy to query blockchain data from any MCP-compatible client.\n\n## For LLM Users: SQL Query Workflow Guide\n\nWhen using this MCP server to run SQL queries on blockchain data, follow this workflow:\n\n1. **Download data** with `query_dataset`:\n ```python\n result = query_dataset(\n dataset=\"blocks\", # or \"transactions\", \"logs\", etc.\n blocks=\"15000000:15001000\", # or use blocks_from_latest=100\n output_format=\"parquet\" # important: use parquet for SQL\n )\n files = result.get(\"files\", []) # Get the returned file paths\n ```\n\n2. **Explore schema** with `get_sql_table_schema`:\n ```python\n # Check what columns are available in the file\n schema = get_sql_table_schema(files[0])\n # Now you can see all columns, data types, and sample data\n ```\n\n3. **Run SQL** with `query_sql`:\n ```python\n # Option 1: Simple table reference (DuckDB will match the table name to file)\n sql_result = query_sql(\n query=\"SELECT block_number, timestamp, gas_used FROM blocks\",\n files=files # Pass the files from step 1\n )\n \n # Option 2: Using read_parquet() with explicit file path\n sql_result = query_sql(\n query=f\"SELECT block_number, timestamp, gas_used FROM read_parquet('{files[0]}')\",\n files=files # Pass the files from step 1\n )\n ```\n\nAlternatively, use the combined approach with `query_blockchain_sql`:\n```python\n# Option 1: Simple table reference\nresult = query_blockchain_sql(\n sql_query=\"SELECT * FROM blocks\",\n dataset=\"blocks\",\n blocks_from_latest=100\n)\n\n# Option 2: Using read_parquet()\nresult = query_blockchain_sql(\n sql_query=\"SELECT * FROM read_parquet('/path/to/file.parquet')\", # Path doesn't matter\n dataset=\"blocks\",\n blocks_from_latest=100\n)\n```\n\nFor a complete working example, see [examples/sql_workflow_example.py](examples/sql_workflow_example.py).\n\n## Features\n\n- **Full Cryo Dataset Access**: Query any Cryo dataset through an API server\n- **MCP Integration**: Works seamlessly with MCP clients\n- **Flexible Query Options**: Support for all major Cryo filtering and output options\n- **Block Range Options**: Query specific blocks, latest block, or relative ranges\n- **Contract Filtering**: Filter data by contract address \n- **Latest Block Access**: Easy access to the latest Ethereum block data\n- **Multiple Output Formats**: JSON, CSV, and Parquet support\n- **Schema Information**: Get detailed dataset schemas and sample data\n- **SQL Queries**: Run SQL queries directly against downloaded blockchain data\n\n## Installation (Optional)\n\nThis is not required if you will run the tool with `uvx` directly.\n\n```bash\n# install with UV (recommended)\nuv tool install cryo-mcp\n```\n\n## Requirements\n\n- Python 3.8+\n- uv\n- A working installation of [Cryo](https://github.com/paradigmxyz/cryo)\n- Access to an Ethereum RPC endpoint\n- DuckDB (for SQL query functionality)\n\n## Quick Start\n\n### Usage with Claude Code\n\n1. Run `claude mcp add` for an interactive prompt.\n2. Enter `uvx` as the command to run.\n3. Enter `cryo-mcp --rpc-url [--data-dir ]` as the args\n4. Alternatively, provide `ETH_RPC_URL` and `CRYO_DATA_DIR` as environment variables instead.\n\nNew instances of `claude` will now have access to cryo as configured to hit your RPC endpoint and store data in the specified directory.\n\n## Available Tools\n\nCryo MCP exposes the following MCP tools:\n\n### `list_datasets()`\n\nReturns a list of all available Cryo datasets.\n\nExample:\n```python\nclient.list_datasets()\n```\n\n### `query_dataset()`\n\nQuery a Cryo dataset with various filtering options.\n\nParameters:\n- `dataset` (str): The name of the dataset to query (e.g., 'blocks', 'transactions', 'logs')\n- `blocks` (str, optional): Block range specification (e.g., '1000:1010')\n- `start_block` (int, optional): Start block number (alternative to blocks)\n- `end_block` (int, optional): End block number (alternative to blocks)\n- `use_latest` (bool, optional): If True, query the latest block\n- `blocks_from_latest` (int, optional): Number of blocks from latest to include\n- `contract` (str, optional): Contract address to filter by\n- `output_format` (str, optional): Output format ('json', 'csv', 'parquet')\n- `include_columns` (list, optional): Columns to include alongside defaults\n- `exclude_columns` (list, optional): Columns to exclude from defaults\n\nExample:\n```python\n# Get transactions from blocks 15M to 15.01M\nclient.query_dataset('transactions', blocks='15M:15.01M')\n\n# Get logs for a specific contract from the latest 100 blocks\nclient.query_dataset('logs', blocks_from_latest=100, contract='0x1234...')\n\n# Get just the latest block\nclient.query_dataset('blocks', use_latest=True)\n```\n\n### `lookup_dataset()`\n\nGet detailed information about a specific dataset, including schema and sample data.\n\nParameters:\n- `name` (str): The name of the dataset to look up\n- `sample_start_block` (int, optional): Start block for sample data\n- `sample_end_block` (int, optional): End block for sample data\n- `use_latest_sample` (bool, optional): Use latest block for sample\n- `sample_blocks_from_latest` (int, optional): Number of blocks from latest for sample\n\nExample:\n```python\nclient.lookup_dataset('logs')\n```\n\n### `get_latest_ethereum_block()`\n\nReturns information about the latest Ethereum block.\n\nExample:\n```python\nclient.get_latest_ethereum_block()\n```\n\n### SQL Query Tools\n\nCryo MCP includes several tools for running SQL queries against blockchain data:\n\n### `query_sql()`\n\nRun a SQL query against downloaded blockchain data.\n\nParameters:\n- `query` (str): SQL query to execute\n- `files` (list, optional): List of parquet file paths to query. If None, will use all files in the data directory.\n- `include_schema` (bool, optional): Whether to include schema information in the result\n\nExample:\n```python\n# Run against all available files\nclient.query_sql(\"SELECT * FROM read_parquet('/path/to/blocks.parquet') LIMIT 10\")\n\n# Run against specific files\nclient.query_sql(\n \"SELECT * FROM read_parquet('/path/to/blocks.parquet') LIMIT 10\",\n files=['/path/to/blocks.parquet']\n)\n```\n\n### `query_blockchain_sql()`\n\nQuery blockchain data using SQL, automatically downloading any required data.\n\nParameters:\n- `sql_query` (str): SQL query to execute\n- `dataset` (str, optional): The dataset to query (e.g., 'blocks', 'transactions')\n- `blocks` (str, optional): Block range specification\n- `start_block` (int, optional): Start block number\n- `end_block` (int, optional): End block number\n- `use_latest` (bool, optional): If True, query the latest block\n- `blocks_from_latest` (int, optional): Number of blocks before the latest to include\n- `contract` (str, optional): Contract address to filter by\n- `force_refresh` (bool, optional): Force download of new data even if it exists\n- `include_schema` (bool, optional): Include schema information in the result\n\nExample:\n```python\n# Automatically downloads blocks data if needed, then runs the SQL query\nclient.query_blockchain_sql(\n sql_query=\"SELECT block_number, gas_used, timestamp FROM blocks ORDER BY gas_used DESC LIMIT 10\",\n dataset=\"blocks\",\n blocks_from_latest=100\n)\n```\n\n### `list_available_sql_tables()`\n\nList all available tables that can be queried with SQL.\n\nExample:\n```python\nclient.list_available_sql_tables()\n```\n\n### `get_sql_table_schema()`\n\nGet the schema for a specific parquet file.\n\nParameters:\n- `file_path` (str): Path to the parquet file\n\nExample:\n```python\nclient.get_sql_table_schema(\"/path/to/blocks.parquet\")\n```\n\n### `get_sql_examples()`\n\nGet example SQL queries for different blockchain datasets.\n\nExample:\n```python\nclient.get_sql_examples()\n```\n\n## Configuration Options\n\nWhen starting the Cryo MCP server, you can use these command-line options:\n\n- `--rpc-url URL`: Ethereum RPC URL (overrides ETH_RPC_URL environment variable)\n- `--data-dir PATH`: Directory to store downloaded data (overrides CRYO_DATA_DIR environment variable, defaults to ~/.cryo-mcp/data/)\n\n## Environment Variables\n\n- `ETH_RPC_URL`: Default Ethereum RPC URL to use when not specified via command line\n- `CRYO_DATA_DIR`: Default directory to store downloaded data when not specified via command line\n\n## Advanced Usage\n\n### SQL Queries Against Blockchain Data\n\nCryo MCP allows you to run powerful SQL queries against blockchain data, combining the flexibility of SQL with Cryo's data extraction capabilities:\n\n#### Two-Step SQL Query Flow\n\nYou can split data extraction and querying into two separate steps:\n\n```python\n# Step 1: Download data and get file paths\ndownload_result = client.query_dataset(\n dataset=\"transactions\",\n blocks_from_latest=1000,\n output_format=\"parquet\"\n)\n\n# Step 2: Use the file paths to run SQL queries\nfile_paths = download_result.get(\"files\", [])\nclient.query_sql(\n query=f\"\"\"\n SELECT \n to_address as contract_address, \n COUNT(*) as tx_count,\n SUM(gas_used) as total_gas,\n AVG(gas_used) as avg_gas\n FROM read_parquet('{file_paths[0]}')\n WHERE to_address IS NOT NULL\n GROUP BY to_address\n ORDER BY total_gas DESC\n LIMIT 20\n \"\"\",\n files=file_paths\n)\n```\n\n#### Combined SQL Query Flow\n\nFor convenience, you can also use the combined function that handles both steps:\n\n```python\n# Get top gas-consuming contracts\nclient.query_blockchain_sql(\n sql_query=\"\"\"\n SELECT \n to_address as contract_address, \n COUNT(*) as tx_count,\n SUM(gas_used) as total_gas,\n AVG(gas_used) as avg_gas\n FROM read_parquet('/path/to/transactions.parquet')\n WHERE to_address IS NOT NULL\n GROUP BY to_address\n ORDER BY total_gas DESC\n LIMIT 20\n \"\"\",\n dataset=\"transactions\",\n blocks_from_latest=1000\n)\n\n# Find blocks with the most transactions\nclient.query_blockchain_sql(\n sql_query=\"\"\"\n SELECT \n block_number, \n COUNT(*) as tx_count\n FROM read_parquet('/path/to/transactions.parquet')\n GROUP BY block_number\n ORDER BY tx_count DESC\n LIMIT 10\n \"\"\",\n dataset=\"transactions\",\n blocks=\"15M:16M\"\n)\n\n# Analyze event logs by topic\nclient.query_blockchain_sql(\n sql_query=\"\"\"\n SELECT \n topic0, \n COUNT(*) as event_count\n FROM read_parquet('/path/to/logs.parquet')\n GROUP BY topic0\n ORDER BY event_count DESC\n LIMIT 20\n \"\"\",\n dataset=\"logs\",\n blocks_from_latest=100\n)\n```\n\n**Note**: For SQL queries, always use `output_format=\"parquet\"` when downloading data to ensure optimal performance with DuckDB. When using `query_blockchain_sql`, you should refer to the file paths directly in your SQL using the `read_parquet()` function.\n\n### Querying with Block Ranges\n\nCryo MCP supports the full range of Cryo's block specification syntax:\n\n```python\n# Using block numbers\nclient.query_dataset('transactions', blocks='15000000:15001000')\n\n# Using K/M notation\nclient.query_dataset('logs', blocks='15M:15.01M')\n\n# Using offsets from latest \nclient.query_dataset('blocks', blocks_from_latest=100)\n```\n\n### Contract Filtering\n\nFilter logs and other data by contract address:\n\n```python\n# Get all logs for USDC contract\nclient.query_dataset('logs', \n blocks='16M:16.1M', \n contract='0xa0b86991c6218b36c1d19d4a2e9eb0ce3606eb48')\n```\n\n### Column Selection\n\nInclude only the columns you need:\n\n```python\n# Get just block numbers and timestamps\nclient.query_dataset('blocks', \n blocks='16M:16.1M', \n include_columns=['number', 'timestamp'])\n```\n\n## Development\n\n### Project Structure\n\n```\ncryo-mcp/\n├── cryo_mcp/ # Main package directory\n│ ├── __init__.py # Package initialization\n│ ├── server.py # Main MCP server implementation\n│ ├── sql.py # SQL query functionality\n├── tests/ # Test directory\n│ ├── test_*.py # Test files\n├── pyproject.toml # Project configuration\n├── README.md # Project documentation\n```\n\n### Run Tests\n\n`uv run pytest`\n\n## License\n\nMIT\n\n## Credits\n\n- Built on top of the amazing [Cryo](https://github.com/paradigmxyz/cryo) tool by Paradigm\n- Uses the [MCP protocol](https://github.com/mcp-team/mcp) for API communication\n","English","tag","en-US",{"_29":41,"_33":45,"_38":46},{"_31":42,"_33":43,"_35":44},"Un servidor API que implementa el Protocolo de finalización de modelos (MCP) para la extracción de datos de la cadena de bloques Cryo, lo que permite a los usuarios consultar datos de la cadena de bloques Ethereum a través de cualquier cliente compatible con MCP.","Servidor Cryo MCP","# Crio MCP 🧊\n\nUn servidor de Protocolo de Finalización de Modelos (MCP) para la herramienta de extracción de datos de blockchain [Cryo](https://github.com/paradigmxyz/cryo) .\n\nCryo MCP le permite acceder a las poderosas capacidades de extracción de datos de blockchain de Cryo a través de un servidor API que implementa el protocolo MCP, lo que facilita la consulta de datos de blockchain desde cualquier cliente compatible con MCP.\n\n## Para usuarios de LLM: Guía de flujo de trabajo de consultas SQL\n\nAl utilizar este servidor MCP para ejecutar consultas SQL en datos de blockchain, siga este flujo de trabajo:\n\n1. **Descargar datos** con `query_dataset` :\n\n ```python\n result = query_dataset(\n dataset=\"blocks\", # or \"transactions\", \"logs\", etc.\n blocks=\"15000000:15001000\", # or use blocks_from_latest=100\n output_format=\"parquet\" # important: use parquet for SQL\n )\n files = result.get(\"files\", []) # Get the returned file paths\n ```\n\n2. **Explorar el esquema** con `get_sql_table_schema` :\n\n ```python\n # Check what columns are available in the file\n schema = get_sql_table_schema(files[0])\n # Now you can see all columns, data types, and sample data\n ```\n\n3. **Ejecutar SQL** con `query_sql` :\n\n ```python\n # Option 1: Simple table reference (DuckDB will match the table name to file)\n sql_result = query_sql(\n query=\"SELECT block_number, timestamp, gas_used FROM blocks\",\n files=files # Pass the files from step 1\n )\n\n # Option 2: Using read_parquet() with explicit file path\n sql_result = query_sql(\n query=f\"SELECT block_number, timestamp, gas_used FROM read_parquet('{files[0]}')\",\n files=files # Pass the files from step 1\n )\n ```\n\nAlternativamente, utilice el enfoque combinado con `query_blockchain_sql` :\n\n```python\n# Option 1: Simple table reference\nresult = query_blockchain_sql(\n sql_query=\"SELECT * FROM blocks\",\n dataset=\"blocks\",\n blocks_from_latest=100\n)\n\n# Option 2: Using read_parquet()\nresult = query_blockchain_sql(\n sql_query=\"SELECT * FROM read_parquet('/path/to/file.parquet')\", # Path doesn't matter\n dataset=\"blocks\",\n blocks_from_latest=100\n)\n```\n\nPara obtener un ejemplo completo y funcional, consulte [examples/sql\\_workflow\\_example.py](examples/sql_workflow_example.py) .\n\n## Características\n\n* **Acceso completo al conjunto de datos criogénicos** : consulte cualquier conjunto de datos criogénicos a través de un servidor API\n* **Integración con MCP** : funciona a la perfección con los clientes de MCP\n* **Opciones de consulta flexibles** : compatibilidad con todas las principales opciones de filtrado y salida de Cryo\n* **Opciones de rango de bloques** : consulta bloques específicos, el último bloque o rangos relativos\n* **Filtrado de contratos** : Filtrar datos por dirección del contrato\n* **Acceso al último bloque** : fácil acceso a los últimos datos del bloque de Ethereum\n* **Múltiples formatos de salida** : compatibilidad con JSON, CSV y Parquet\n* **Información del esquema** : obtenga esquemas de conjuntos de datos detallados y datos de muestra\n* **Consultas SQL** : ejecute consultas SQL directamente contra los datos de blockchain descargados\n\n## Instalación (opcional)\n\nEsto no es necesario si va a ejecutar la herramienta con `uvx` directamente.\n\n```bash\n# install with UV (recommended)\nuv tool install cryo-mcp\n```\n\n## Requisitos\n\n* Python 3.8+\n* ultravioleta\n* Una instalación de [Cryo](https://github.com/paradigmxyz/cryo) en funcionamiento\n* Acceso a un punto final de Ethereum RPC\n* DuckDB (para la funcionalidad de consulta SQL)\n\n## Inicio rápido\n\n### Uso con Claude Code\n\n1. Ejecute `claude mcp add` para obtener un mensaje interactivo.\n2. Introduzca `uvx` como comando a ejecutar.\n3. Ingrese `cryo-mcp --rpc-url [--data-dir ]` como argumentos\n4. Como alternativa, proporcione `ETH_RPC_URL` y `CRYO_DATA_DIR` como variables de entorno.\n\nLas nuevas instancias de `claude` ahora tendrán acceso a cryo según lo configurado para llegar a su punto final RPC y almacenar datos en el directorio especificado.\n\n## Herramientas disponibles\n\nCryo MCP expone las siguientes herramientas MCP:\n\n### `list_datasets()`\n\nDevuelve una lista de todos los conjuntos de datos Cryo disponibles.\n\nEjemplo:\n\n```python\nclient.list_datasets()\n```\n\n### `query_dataset()`\n\nConsulta un conjunto de datos criogénicos con varias opciones de filtrado.\n\nParámetros:\n\n* `dataset` (str): el nombre del conjunto de datos a consultar (por ejemplo, 'bloques', 'transacciones', 'registros')\n* `blocks` (str, opcional): Especificación del rango de bloques (por ejemplo, '1000:1010')\n* `start_block` (int, opcional): Número de bloque de inicio (alternativa a bloques)\n* `end_block` (int, opcional): Número de bloque final (alternativa a bloques)\n* `use_latest` (bool, opcional): si es verdadero, consulta el último bloque\n* `blocks_from_latest` (int, opcional): Número de bloques desde el último a incluir\n* `contract` (str, opcional): Dirección del contrato por la que filtrar\n* `output_format` (str, opcional): Formato de salida ('json', 'csv', 'parquet')\n* `include_columns` (lista, opcional): columnas que se incluirán junto con los valores predeterminados\n* `exclude_columns` (lista, opcional): columnas para excluir de los valores predeterminados\n\nEjemplo:\n\n```python\n# Get transactions from blocks 15M to 15.01M\nclient.query_dataset('transactions', blocks='15M:15.01M')\n\n# Get logs for a specific contract from the latest 100 blocks\nclient.query_dataset('logs', blocks_from_latest=100, contract='0x1234...')\n\n# Get just the latest block\nclient.query_dataset('blocks', use_latest=True)\n```\n\n### `lookup_dataset()`\n\nObtenga información detallada sobre un conjunto de datos específico, incluido el esquema y los datos de muestra.\n\nParámetros:\n\n* `name` (str): El nombre del conjunto de datos a buscar\n* `sample_start_block` (int, opcional): bloque de inicio para datos de muestra\n* `sample_end_block` (int, opcional): Fin del bloque para datos de muestra\n* `use_latest_sample` (bool, opcional): utiliza el último bloque para la muestra\n* `sample_blocks_from_latest` (int, opcional): Número de bloques desde el último para la muestra\n\nEjemplo:\n\n```python\nclient.lookup_dataset('logs')\n```\n\n### `get_latest_ethereum_block()`\n\nDevuelve información sobre el último bloque de Ethereum.\n\nEjemplo:\n\n```python\nclient.get_latest_ethereum_block()\n```\n\n### Herramientas de consulta SQL\n\nCryo MCP incluye varias herramientas para ejecutar consultas SQL en datos de blockchain:\n\n### `query_sql()`\n\nEjecute una consulta SQL contra los datos de blockchain descargados.\n\nParámetros:\n\n* `query` (str): consulta SQL a ejecutar\n* `files` (lista, opcional): Lista de rutas de archivos de Parquet para consultar. Si no hay ninguna, se usarán todos los archivos del directorio de datos.\n* `include_schema` (bool, opcional): si se debe incluir información del esquema en el resultado\n\nEjemplo:\n\n```python\n# Run against all available files\nclient.query_sql(\"SELECT * FROM read_parquet('/path/to/blocks.parquet') LIMIT 10\")\n\n# Run against specific files\nclient.query_sql(\n \"SELECT * FROM read_parquet('/path/to/blocks.parquet') LIMIT 10\",\n files=['/path/to/blocks.parquet']\n)\n```\n\n### `query_blockchain_sql()`\n\nConsulta datos de blockchain usando SQL y descarga automáticamente cualquier dato requerido.\n\nParámetros:\n\n* `sql_query` (str): consulta SQL a ejecutar\n* `dataset` (str, opcional): el conjunto de datos a consultar (por ejemplo, 'bloques', 'transacciones')\n* `blocks` (str, opcional): Especificación del rango de bloques\n* `start_block` (int, opcional): Número de bloque de inicio\n* `end_block` (int, opcional): Número de bloque final\n* `use_latest` (bool, opcional): si es verdadero, consulta el último bloque\n* `blocks_from_latest` (int, opcional): Número de bloques antes del último a incluir\n* `contract` (str, opcional): Dirección del contrato por la que filtrar\n* `force_refresh` (bool, opcional): fuerza la descarga de nuevos datos incluso si existen\n* `include_schema` (bool, opcional): incluye información del esquema en el resultado\n\nEjemplo:\n\n```python\n# Automatically downloads blocks data if needed, then runs the SQL query\nclient.query_blockchain_sql(\n sql_query=\"SELECT block_number, gas_used, timestamp FROM blocks ORDER BY gas_used DESC LIMIT 10\",\n dataset=\"blocks\",\n blocks_from_latest=100\n)\n```\n\n### `list_available_sql_tables()`\n\nEnumere todas las tablas disponibles que se pueden consultar con SQL.\n\nEjemplo:\n\n```python\nclient.list_available_sql_tables()\n```\n\n### `get_sql_table_schema()`\n\nObtenga el esquema para un archivo parquet específico.\n\nParámetros:\n\n* `file_path` (str): Ruta al archivo parquet\n\nEjemplo:\n\n```python\nclient.get_sql_table_schema(\"/path/to/blocks.parquet\")\n```\n\n### `get_sql_examples()`\n\nObtenga ejemplos de consultas SQL para diferentes conjuntos de datos de blockchain.\n\nEjemplo:\n\n```python\nclient.get_sql_examples()\n```\n\n## Opciones de configuración\n\nAl iniciar el servidor Cryo MCP, puede utilizar estas opciones de línea de comandos:\n\n* `--rpc-url URL` : URL de RPC de Ethereum (anula la variable de entorno ETH\\_RPC\\_URL)\n* `--data-dir PATH` : Directorio para almacenar los datos descargados (anula la variable de entorno CRYO\\_DATA\\_DIR, el valor predeterminado es ~/.cryo-mcp/data/)\n\n## Variables de entorno\n\n* `ETH_RPC_URL` : URL de RPC de Ethereum predeterminada que se utilizará cuando no se especifique mediante la línea de comando\n* `CRYO_DATA_DIR` : Directorio predeterminado para almacenar los datos descargados cuando no se especifica mediante la línea de comando\n\n## Uso avanzado\n\n### Consultas SQL contra datos de blockchain\n\nCryo MCP le permite ejecutar potentes consultas SQL contra datos de blockchain, combinando la flexibilidad de SQL con las capacidades de extracción de datos de Cryo:\n\n#### Flujo de consulta SQL de dos pasos\n\nPuede dividir la extracción y consulta de datos en dos pasos separados:\n\n```python\n# Step 1: Download data and get file paths\ndownload_result = client.query_dataset(\n dataset=\"transactions\",\n blocks_from_latest=1000,\n output_format=\"parquet\"\n)\n\n# Step 2: Use the file paths to run SQL queries\nfile_paths = download_result.get(\"files\", [])\nclient.query_sql(\n query=f\"\"\"\n SELECT \n to_address as contract_address, \n COUNT(*) as tx_count,\n SUM(gas_used) as total_gas,\n AVG(gas_used) as avg_gas\n FROM read_parquet('{file_paths[0]}')\n WHERE to_address IS NOT NULL\n GROUP BY to_address\n ORDER BY total_gas DESC\n LIMIT 20\n \"\"\",\n files=file_paths\n)\n```\n\n#### Flujo de consulta SQL combinado\n\nPara mayor comodidad, también puede utilizar la función combinada que maneja ambos pasos:\n\n```python\n# Get top gas-consuming contracts\nclient.query_blockchain_sql(\n sql_query=\"\"\"\n SELECT \n to_address as contract_address, \n COUNT(*) as tx_count,\n SUM(gas_used) as total_gas,\n AVG(gas_used) as avg_gas\n FROM read_parquet('/path/to/transactions.parquet')\n WHERE to_address IS NOT NULL\n GROUP BY to_address\n ORDER BY total_gas DESC\n LIMIT 20\n \"\"\",\n dataset=\"transactions\",\n blocks_from_latest=1000\n)\n\n# Find blocks with the most transactions\nclient.query_blockchain_sql(\n sql_query=\"\"\"\n SELECT \n block_number, \n COUNT(*) as tx_count\n FROM read_parquet('/path/to/transactions.parquet')\n GROUP BY block_number\n ORDER BY tx_count DESC\n LIMIT 10\n \"\"\",\n dataset=\"transactions\",\n blocks=\"15M:16M\"\n)\n\n# Analyze event logs by topic\nclient.query_blockchain_sql(\n sql_query=\"\"\"\n SELECT \n topic0, \n COUNT(*) as event_count\n FROM read_parquet('/path/to/logs.parquet')\n GROUP BY topic0\n ORDER BY event_count DESC\n LIMIT 20\n \"\"\",\n dataset=\"logs\",\n blocks_from_latest=100\n)\n```\n\n**Nota** : Para consultas SQL, utilice siempre `output_format=\"parquet\"` al descargar datos para garantizar un rendimiento óptimo con DuckDB. Al usar `query_blockchain_sql` , consulte las rutas de archivo directamente en su SQL mediante la función `read_parquet()` .\n\n### Consultas con rangos de bloques\n\nCryo MCP admite toda la sintaxis de especificación de bloques de Cryo:\n\n```python\n# Using block numbers\nclient.query_dataset('transactions', blocks='15000000:15001000')\n\n# Using K/M notation\nclient.query_dataset('logs', blocks='15M:15.01M')\n\n# Using offsets from latest \nclient.query_dataset('blocks', blocks_from_latest=100)\n```\n\n### Filtrado de contratos\n\nFiltrar registros y otros datos por dirección de contrato:\n\n```python\n# Get all logs for USDC contract\nclient.query_dataset('logs', \n blocks='16M:16.1M', \n contract='0xa0b86991c6218b36c1d19d4a2e9eb0ce3606eb48')\n```\n\n### Selección de columnas\n\nIncluya sólo las columnas que necesita:\n\n```python\n# Get just block numbers and timestamps\nclient.query_dataset('blocks', \n blocks='16M:16.1M', \n include_columns=['number', 'timestamp'])\n```\n\n## Desarrollo\n\n### Estructura del proyecto\n\n```\ncryo-mcp/\n├── cryo_mcp/ # Main package directory\n│ ├── __init__.py # Package initialization\n│ ├── server.py # Main MCP server implementation\n│ ├── sql.py # SQL query functionality\n├── tests/ # Test directory\n│ ├── test_*.py # Test files\n├── pyproject.toml # Project configuration\n├── README.md # Project documentation\n```\n\n### Ejecutar pruebas\n\n`uv run pytest`\n\n## Licencia\n\nInstituto Tecnológico de Massachusetts (MIT)\n\n## Créditos\n\n* Construido sobre la increíble herramienta [Cryo](https://github.com/paradigmxyz/cryo) de Paradigm\n* Utiliza el [protocolo MCP](https://github.com/mcp-team/mcp) para la comunicación API","Spanish","es-ES",{"_29":48,"_33":52,"_38":53},{"_31":49,"_33":50,"_35":51},"Cryo ブロックチェーンデータ抽出用のモデル補完プロトコル (MCP) を実装する API サーバー。ユーザーは、MCP 互換のクライアントを通じて Ethereum ブロックチェーンデータを照会できます。","Cryo MCP サーバー","# クライオMCP🧊\n\n[Cryo](https://github.com/paradigmxyz/cryo)ブロックチェーンデータ抽出ツール用のモデル完了プロトコル (MCP) サーバー。\n\nCryo MCP を使用すると、MCP プロトコルを実装する API サーバーを介して Cryo の強力なブロックチェーンデータ抽出機能にアクセスできるため、MCP 対応のクライアントからブロックチェーンデータを簡単に照会できます。\n\n## LLM ユーザー向け: SQL クエリワークフローガイド\n\nこの MCP サーバーを使用してブロックチェーンデータに対して SQL クエリを実行する場合は、次のワークフローに従います。\n\n1. `query_dataset`を使用して**データをダウンロードします**。\n\n ```python\n result = query_dataset(\n dataset=\"blocks\", # or \"transactions\", \"logs\", etc.\n blocks=\"15000000:15001000\", # or use blocks_from_latest=100\n output_format=\"parquet\" # important: use parquet for SQL\n )\n files = result.get(\"files\", []) # Get the returned file paths\n ```\n\n2. `get_sql_table_schema`を使用して**スキーマを探索します**。\n\n ```python\n # Check what columns are available in the file\n schema = get_sql_table_schema(files[0])\n # Now you can see all columns, data types, and sample data\n ```\n\n3. `query_sql`で**SQL を実行します**。\n\n ```python\n # Option 1: Simple table reference (DuckDB will match the table name to file)\n sql_result = query_sql(\n query=\"SELECT block_number, timestamp, gas_used FROM blocks\",\n files=files # Pass the files from step 1\n )\n\n # Option 2: Using read_parquet() with explicit file path\n sql_result = query_sql(\n query=f\"SELECT block_number, timestamp, gas_used FROM read_parquet('{files[0]}')\",\n files=files # Pass the files from step 1\n )\n ```\n\nあるいは、 `query_blockchain_sql`と組み合わせたアプローチを使用します。\n\n```python\n# Option 1: Simple table reference\nresult = query_blockchain_sql(\n sql_query=\"SELECT * FROM blocks\",\n dataset=\"blocks\",\n blocks_from_latest=100\n)\n\n# Option 2: Using read_parquet()\nresult = query_blockchain_sql(\n sql_query=\"SELECT * FROM read_parquet('/path/to/file.parquet')\", # Path doesn't matter\n dataset=\"blocks\",\n blocks_from_latest=100\n)\n```\n\n完全な動作例については、 [examples/sql\\_workflow\\_example.py](examples/sql_workflow_example.py)を参照してください。\n\n## 特徴\n\n* **完全なCryoデータセットアクセス**：APIサーバーを介して任意のCryoデータセットをクエリします\n* **MCP統合**:MCPクライアントとシームレスに連携\n* **柔軟なクエリオプション**: すべての主要なCryoフィルタリングと出力オプションをサポート\n* **ブロック範囲オプション**: 特定のブロック、最新のブロック、または相対範囲を照会します\n* **契約フィルタリング**: 契約アドレスでデータをフィルタリング\n* **最新ブロックアクセス**：最新のイーサリアムブロックデータに簡単にアクセスできます\n* **複数の出力形式**: JSON、CSV、Parquet をサポート\n* **スキーマ情報**: 詳細なデータセットスキーマとサンプルデータを取得します\n* **SQLクエリ**:ダウンロードしたブロックチェーンデータに対して直接SQLクエリを実行します。\n\n## インストール（オプション）\n\n`uvx`を使用してツールを直接実行する場合、これは必要ありません。\n\n```bash\n# install with UV (recommended)\nuv tool install cryo-mcp\n```\n\n## 要件\n\n* Python 3.8以上\n* 紫外線\n* [Cryo](https://github.com/paradigmxyz/cryo)の稼働中の設備\n* Ethereum RPCエンドポイントへのアクセス\n* DuckDB（SQLクエリ機能用）\n\n## クイックスタート\n\n### クロードコードでの使用\n\n1. 対話型プロンプトを表示するには`claude mcp add`実行します。\n2. 実行するコマンドとして`uvx`と入力します。\n3. 引数として`cryo-mcp --rpc-url [--data-dir ]`を入力します。\n4. あるいは、代わりに`ETH_RPC_URL`と`CRYO_DATA_DIR`環境変数として指定します。\n\n`claude`の新しいインスタンスは、RPC エンドポイントにアクセスし、指定されたディレクトリにデータを保存するように構成されたとおりに cryo にアクセスできるようになっています。\n\n## 利用可能なツール\n\nCryo MCP は次の MCP ツールを公開します。\n\n### `list_datasets()`\n\n利用可能なすべての Cryo データセットのリストを返します。\n\n例：\n\n```python\nclient.list_datasets()\n```\n\n### `query_dataset()`\n\nさまざまなフィルタリングオプションを使用して Cryo データセットをクエリします。\n\nパラメータ:\n\n* `dataset` (str): クエリするデータセットの名前 (例: 'blocks'、'transactions'、'logs')\n* `blocks` （文字列、オプション）: ブロック範囲の指定（例：'1000:1010'）\n* `start_block` (int, オプション): 開始ブロック番号（blocks の代替）\n* `end_block` (int, オプション): 終了ブロック番号 (blocks の代替)\n* `use_latest` (bool, オプション): Trueの場合、最新のブロックを照会します\n* `blocks_from_latest` (int, オプション): 最新から含めるブロックの数\n* `contract` (文字列、オプション): フィルタリングする契約アドレス\n* `output_format` (str, オプション): 出力形式 ('json', 'csv', 'parquet')\n* `include_columns` (リスト、オプション): デフォルトと一緒に含める列\n* `exclude_columns` (リスト、オプション): デフォルトから除外する列\n\n例：\n\n```python\n# Get transactions from blocks 15M to 15.01M\nclient.query_dataset('transactions', blocks='15M:15.01M')\n\n# Get logs for a specific contract from the latest 100 blocks\nclient.query_dataset('logs', blocks_from_latest=100, contract='0x1234...')\n\n# Get just the latest block\nclient.query_dataset('blocks', use_latest=True)\n```\n\n### `lookup_dataset()`\n\nスキーマやサンプルデータなど、特定のデータセットに関する詳細情報を取得します。\n\nパラメータ:\n\n* `name` (str): 検索するデータセットの名前\n* `sample_start_block` (int, オプション): サンプルデータの開始ブロック\n* `sample_end_block` (int, オプション): サンプルデータの終了ブロック\n* `use_latest_sample` (bool, オプション): サンプルに最新のブロックを使用する\n* `sample_blocks_from_latest` (int, オプション): サンプルの最新からのブロック数\n\n例：\n\n```python\nclient.lookup_dataset('logs')\n```\n\n### `get_latest_ethereum_block()`\n\n最新の Ethereum ブロックに関する情報を返します。\n\n例：\n\n```python\nclient.get_latest_ethereum_block()\n```\n\n### SQLクエリツール\n\nCryo MCP には、ブロックチェーンデータに対して SQL クエリを実行するためのツールがいくつか含まれています。\n\n### `query_sql()`\n\nダウンロードしたブロックチェーンデータに対して SQL クエリを実行します。\n\nパラメータ:\n\n* `query` (str): 実行するSQLクエリ\n* `files` (リスト, オプション): クエリ対象の parquet ファイルパスのリスト。指定しない場合は、データディレクトリ内のすべてのファイルが使用されます。\n* `include_schema` (bool, オプション): 結果にスキーマ情報を含めるかどうか\n\n例：\n\n```python\n# Run against all available files\nclient.query_sql(\"SELECT * FROM read_parquet('/path/to/blocks.parquet') LIMIT 10\")\n\n# Run against specific files\nclient.query_sql(\n \"SELECT * FROM read_parquet('/path/to/blocks.parquet') LIMIT 10\",\n files=['/path/to/blocks.parquet']\n)\n```\n\n### `query_blockchain_sql()`\n\nSQL を使用してブロックチェーンデータをクエリし、必要なデータを自動的にダウンロードします。\n\nパラメータ:\n\n* `sql_query` (str): 実行するSQLクエリ\n* `dataset` （文字列、オプション）: クエリするデータセット（例：'blocks'、'transactions'）\n* `blocks` （文字列、オプション）: ブロック範囲の指定\n* `start_block` (int, オプション): 開始ブロック番号\n* `end_block` (int, オプション): 終了ブロック番号\n* `use_latest` (bool, オプション): Trueの場合、最新のブロックを照会します\n* `blocks_from_latest` (int, オプション): 含める最新のブロックの前のブロック数\n* `contract` (文字列、オプション): フィルタリングする契約アドレス\n* `force_refresh` (bool, オプション): 新しいデータが存在する場合でも強制的にダウンロードする\n* `include_schema` (bool, オプション): 結果にスキーマ情報を含める\n\n例：\n\n```python\n# Automatically downloads blocks data if needed, then runs the SQL query\nclient.query_blockchain_sql(\n sql_query=\"SELECT block_number, gas_used, timestamp FROM blocks ORDER BY gas_used DESC LIMIT 10\",\n dataset=\"blocks\",\n blocks_from_latest=100\n)\n```\n\n### `list_available_sql_tables()`\n\nSQL でクエリできる利用可能なすべてのテーブルを一覧表示します。\n\n例：\n\n```python\nclient.list_available_sql_tables()\n```\n\n### `get_sql_table_schema()`\n\n特定の parquet ファイルのスキーマを取得します。\n\nパラメータ:\n\n* `file_path` (str): parquetファイルへのパス\n\n例：\n\n```python\nclient.get_sql_table_schema(\"/path/to/blocks.parquet\")\n```\n\n### `get_sql_examples()`\n\nさまざまなブロックチェーンデータセットの SQL クエリの例を取得します。\n\n例：\n\n```python\nclient.get_sql_examples()\n```\n\n## 設定オプション\n\nCryo MCP サーバーを起動するときに、次のコマンドラインオプションを使用できます。\n\n* `--rpc-url URL` : Ethereum RPC URL (ETH\\_RPC\\_URL 環境変数を上書きします)\n* `--data-dir PATH` : ダウンロードしたデータを保存するディレクトリ (CRYO\\_DATA\\_DIR 環境変数を上書きします。デフォルトは ~/.cryo-mcp/data/ です)\n\n## 環境変数\n\n* `ETH_RPC_URL` : コマンドラインで指定されていない場合に使用するデフォルトの Ethereum RPC URL\n* `CRYO_DATA_DIR` : コマンドラインで指定されていない場合にダウンロードしたデータを保存するデフォルトのディレクトリ\n\n## 高度な使用法\n\n### ブロックチェーンデータに対するSQLクエリ\n\nCryo MCP を使用すると、SQL の柔軟性と Cryo のデータ抽出機能を組み合わせて、ブロックチェーンデータに対して強力な SQL クエリを実行できます。\n\n#### 2段階のSQLクエリフロー\n\nデータの抽出とクエリを 2 つの別々のステップに分割できます。\n\n```python\n# Step 1: Download data and get file paths\ndownload_result = client.query_dataset(\n dataset=\"transactions\",\n blocks_from_latest=1000,\n output_format=\"parquet\"\n)\n\n# Step 2: Use the file paths to run SQL queries\nfile_paths = download_result.get(\"files\", [])\nclient.query_sql(\n query=f\"\"\"\n SELECT \n to_address as contract_address, \n COUNT(*) as tx_count,\n SUM(gas_used) as total_gas,\n AVG(gas_used) as avg_gas\n FROM read_parquet('{file_paths[0]}')\n WHERE to_address IS NOT NULL\n GROUP BY to_address\n ORDER BY total_gas DESC\n LIMIT 20\n \"\"\",\n files=file_paths\n)\n```\n\n#### 複合SQLクエリフロー\n\n便宜上、両方のステップを処理する結合関数を使用することもできます。\n\n```python\n# Get top gas-consuming contracts\nclient.query_blockchain_sql(\n sql_query=\"\"\"\n SELECT \n to_address as contract_address, \n COUNT(*) as tx_count,\n SUM(gas_used) as total_gas,\n AVG(gas_used) as avg_gas\n FROM read_parquet('/path/to/transactions.parquet')\n WHERE to_address IS NOT NULL\n GROUP BY to_address\n ORDER BY total_gas DESC\n LIMIT 20\n \"\"\",\n dataset=\"transactions\",\n blocks_from_latest=1000\n)\n\n# Find blocks with the most transactions\nclient.query_blockchain_sql(\n sql_query=\"\"\"\n SELECT \n block_number, \n COUNT(*) as tx_count\n FROM read_parquet('/path/to/transactions.parquet')\n GROUP BY block_number\n ORDER BY tx_count DESC\n LIMIT 10\n \"\"\",\n dataset=\"transactions\",\n blocks=\"15M:16M\"\n)\n\n# Analyze event logs by topic\nclient.query_blockchain_sql(\n sql_query=\"\"\"\n SELECT \n topic0, \n COUNT(*) as event_count\n FROM read_parquet('/path/to/logs.parquet')\n GROUP BY topic0\n ORDER BY event_count DESC\n LIMIT 20\n \"\"\",\n dataset=\"logs\",\n blocks_from_latest=100\n)\n```\n\n**注**：SQLクエリの場合、DuckDBで最適なパフォーマンスを確保するには、データをダウンロードする際に必ず`output_format=\"parquet\"`を使用してください。query\\_blockchain\\_sql `query_blockchain_sql`使用する場合は、 `read_parquet()`関数を使用してSQL内でファイルパスを直接参照する必要があります。\n\n### ブロック範囲によるクエリ\n\nCryo MCP は、Cryo のブロック仕様構文の全範囲をサポートします。\n\n```python\n# Using block numbers\nclient.query_dataset('transactions', blocks='15000000:15001000')\n\n# Using K/M notation\nclient.query_dataset('logs', blocks='15M:15.01M')\n\n# Using offsets from latest \nclient.query_dataset('blocks', blocks_from_latest=100)\n```\n\n### 契約フィルタリング\n\n契約アドレス別にログやその他のデータをフィルタリングします。\n\n```python\n# Get all logs for USDC contract\nclient.query_dataset('logs', \n blocks='16M:16.1M', \n contract='0xa0b86991c6218b36c1d19d4a2e9eb0ce3606eb48')\n```\n\n### 列の選択\n\n必要な列のみを含めます。\n\n```python\n# Get just block numbers and timestamps\nclient.query_dataset('blocks', \n blocks='16M:16.1M', \n include_columns=['number', 'timestamp'])\n```\n\n## 発達\n\n### プロジェクト構造\n\n```\ncryo-mcp/\n├── cryo_mcp/ # Main package directory\n│ ├── __init__.py # Package initialization\n│ ├── server.py # Main MCP server implementation\n│ ├── sql.py # SQL query functionality\n├── tests/ # Test directory\n│ ├── test_*.py # Test files\n├── pyproject.toml # Project configuration\n├── README.md # Project documentation\n```\n\n### テストを実行する\n\n`uv run pytest`\n\n## ライセンス\n\nマサチューセッツ工科大学\n\n## クレジット\n\n* パラダイムの素晴らしい[Cryo](https://github.com/paradigmxyz/cryo)ツールをベースに構築\n* API通信に[MCPプロトコル](https://github.com/mcp-team/mcp)を使用する","Japanese","ja-JP",{"_29":55,"_33":59,"_38":60},{"_31":56,"_33":57,"_35":58},"Cryo 블록체인 데이터 추출을 위한 모델 완성 프로토콜(MCP)을 구현하는 API 서버로, 사용자는 모든 MCP 호환 클라이언트를 통해 Ethereum 블록체인 데이터를 쿼리할 수 있습니다.","Cryo MCP 서버","# 크라이오 MCP 🧊\n\n[Cryo](https://github.com/paradigmxyz/cryo) 블록체인 데이터 추출 도구를 위한 모델 완성 프로토콜(MCP) 서버입니다.\n\nCryo MCP를 사용하면 MCP 프로토콜을 구현하는 API 서버를 통해 Cryo의 강력한 블록체인 데이터 추출 기능에 액세스할 수 있으므로 모든 MCP 호환 클라이언트에서 블록체인 데이터를 쉽게 쿼리할 수 있습니다.\n\n## LLM 사용자를 위한 SQL 쿼리 워크플로 가이드\n\n이 MCP 서버를 사용하여 블록체인 데이터에 대한 SQL 쿼리를 실행할 때 다음 워크플로를 따르세요.\n\n1. `query_dataset` 사용하여 **데이터 다운로드** :\n\n 지엑스피1\n\n2. `get_sql_table_schema` 로 **스키마를 탐색하세요** :\n\n ```python\n # Check what columns are available in the file\n schema = get_sql_table_schema(files[0])\n # Now you can see all columns, data types, and sample data\n ```\n\n3. `query_sql` 로 **SQL을 실행합니다** :\n\n ```python\n # Option 1: Simple table reference (DuckDB will match the table name to file)\n sql_result = query_sql(\n query=\"SELECT block_number, timestamp, gas_used FROM blocks\",\n files=files # Pass the files from step 1\n )\n\n # Option 2: Using read_parquet() with explicit file path\n sql_result = query_sql(\n query=f\"SELECT block_number, timestamp, gas_used FROM read_parquet('{files[0]}')\",\n files=files # Pass the files from step 1\n )\n ```\n\n또는 `query_blockchain_sql` 과 결합된 접근 방식을 사용하세요.\n\n```python\n# Option 1: Simple table reference\nresult = query_blockchain_sql(\n sql_query=\"SELECT * FROM blocks\",\n dataset=\"blocks\",\n blocks_from_latest=100\n)\n\n# Option 2: Using read_parquet()\nresult = query_blockchain_sql(\n sql_query=\"SELECT * FROM read_parquet('/path/to/file.parquet')\", # Path doesn't matter\n dataset=\"blocks\",\n blocks_from_latest=100\n)\n```\n\n전체 작동 예제를 보려면 [examples/sql\\_workflow\\_example.py를](examples/sql_workflow_example.py) 참조하세요.\n\n## 특징\n\n* **전체 Cryo 데이터 세트 액세스** : API 서버를 통해 모든 Cryo 데이터 세트 쿼리\n* **MCP 통합** : MCP 클라이언트와 원활하게 작동합니다.\n* **유연한 쿼리 옵션** : 모든 주요 Cryo 필터링 및 출력 옵션 지원\n* **블록 범위 옵션** : 특정 블록, 최신 블록 또는 상대 범위 쿼리\n* **계약 필터링** : 계약 주소로 데이터 필터링\n* **최신 블록 접근** : 최신 이더리움 블록 데이터에 쉽게 접근\n* **다양한 출력 형식** : JSON, CSV 및 Parquet 지원\n* **스키마 정보** : 자세한 데이터 세트 스키마 및 샘플 데이터 가져오기\n* **SQL 쿼리** : 다운로드한 블록체인 데이터에 대해 SQL 쿼리를 직접 실행합니다.\n\n## 설치(선택 사항)\n\n`uvx` 로 도구를 직접 실행하는 경우에는 필요하지 않습니다.\n\n```bash\n# install with UV (recommended)\nuv tool install cryo-mcp\n```\n\n## 요구 사항\n\n* 파이썬 3.8 이상\n* 자외선\n* [Cryo](https://github.com/paradigmxyz/cryo) 의 작동 설치\n* Ethereum RPC 엔드포인트에 액세스\n* DuckDB(SQL 쿼리 기능용)\n\n## 빠른 시작\n\n### Claude Code와 함께 사용\n\n1. 대화형 프롬프트를 표시하려면 `claude mcp add` 실행하세요.\n2. 실행할 명령어로 `uvx` 입력합니다.\n3. 인수로 `cryo-mcp --rpc-url [--data-dir ]` 입력하세요.\n4. 또는 `ETH_RPC_URL` 과 `CRYO_DATA_DIR` 환경 변수로 제공하세요.\n\n이제 `claude` 의 새로운 인스턴스는 RPC 엔드포인트에 접근하고 지정된 디렉토리에 데이터를 저장하도록 구성된 Cryo에 액세스할 수 있습니다.\n\n## 사용 가능한 도구\n\nCryo MCP는 다음과 같은 MCP 도구를 제공합니다.\n\n### `list_datasets()`\n\n사용 가능한 모든 Cryo 데이터 세트 목록을 반환합니다.\n\n예:\n\n```python\nclient.list_datasets()\n```\n\n### `query_dataset()`\n\n다양한 필터링 옵션을 사용하여 Cryo 데이터 세트를 쿼리합니다.\n\n매개변수:\n\n* `dataset` (str): 쿼리할 데이터 세트의 이름(예: '블록', '거래', '로그')\n* `blocks` (str, 선택 사항): 블록 범위 지정(예: '1000:1010')\n* `start_block` (int, 선택 사항): 시작 블록 번호(blocks의 대안)\n* `end_block` (int, 선택 사항): 종료 블록 번호(blocks의 대안)\n* `use_latest` (bool, 선택 사항): True인 경우 최신 블록을 쿼리합니다.\n* `blocks_from_latest` (int, 선택 사항): 포함할 최신 블록 수\n* `contract` (str, 선택 사항): 필터링할 계약 주소\n* `output_format` (str, 선택 사항): 출력 형식('json', 'csv', 'parquet')\n* `include_columns` (목록, 선택 사항): 기본값과 함께 포함할 열\n* `exclude_columns` (목록, 선택 사항): 기본값에서 제외할 열\n\n예:\n\n```python\n# Get transactions from blocks 15M to 15.01M\nclient.query_dataset('transactions', blocks='15M:15.01M')\n\n# Get logs for a specific contract from the latest 100 blocks\nclient.query_dataset('logs', blocks_from_latest=100, contract='0x1234...')\n\n# Get just the latest block\nclient.query_dataset('blocks', use_latest=True)\n```\n\n### `lookup_dataset()`\n\n스키마와 샘플 데이터를 포함하여 특정 데이터 세트에 대한 자세한 정보를 얻으세요.\n\n매개변수:\n\n* `name` (str): 조회할 데이터 세트의 이름\n* `sample_start_block` (int, 선택 사항): 샘플 데이터의 시작 블록\n* `sample_end_block` (int, 선택 사항): 샘플 데이터의 종료 블록\n* `use_latest_sample` (bool, 선택 사항): 샘플에 최신 블록을 사용합니다.\n* `sample_blocks_from_latest` (int, 선택 사항): 샘플의 최신 블록 수\n\n예:\n\n```python\nclient.lookup_dataset('logs')\n```\n\n### `get_latest_ethereum_block()`\n\n최신 Ethereum 블록에 대한 정보를 반환합니다.\n\n예:\n\n```python\nclient.get_latest_ethereum_block()\n```\n\n### SQL 쿼리 도구\n\nCryo MCP에는 블록체인 데이터에 대한 SQL 쿼리를 실행하기 위한 여러 도구가 포함되어 있습니다.\n\n### `query_sql()`\n\n다운로드한 블록체인 데이터에 대해 SQL 쿼리를 실행합니다.\n\n매개변수:\n\n* `query` (str): 실행할 SQL 쿼리\n* `files` (목록, 선택 사항): 쿼리할 Parquet 파일 경로 목록입니다. 없으면 데이터 디렉터리의 모든 파일을 사용합니다.\n* `include_schema` (bool, 선택 사항): 결과에 스키마 정보를 포함할지 여부\n\n예:\n\n```python\n# Run against all available files\nclient.query_sql(\"SELECT * FROM read_parquet('/path/to/blocks.parquet') LIMIT 10\")\n\n# Run against specific files\nclient.query_sql(\n \"SELECT * FROM read_parquet('/path/to/blocks.parquet') LIMIT 10\",\n files=['/path/to/blocks.parquet']\n)\n```\n\n### `query_blockchain_sql()`\n\nSQL을 사용하여 블록체인 데이터를 쿼리하고, 필요한 데이터를 자동으로 다운로드합니다.\n\n매개변수:\n\n* `sql_query` (str): 실행할 SQL 쿼리\n* `dataset` (str, 선택 사항): 쿼리할 데이터 세트(예: '블록', '거래')\n* `blocks` (str, 선택 사항): 블록 범위 지정\n* `start_block` (int, 선택 사항): 시작 블록 번호\n* `end_block` (int, 선택 사항): 종료 블록 번호\n* `use_latest` (bool, 선택 사항): True인 경우 최신 블록을 쿼리합니다.\n* `blocks_from_latest` (int, 선택 사항): 포함할 최신 블록 이전의 블록 수\n* `contract` (str, 선택 사항): 필터링할 계약 주소\n* `force_refresh` (bool, 선택 사항): 새 데이터가 존재하더라도 강제로 다운로드합니다.\n* `include_schema` (bool, 선택 사항): 결과에 스키마 정보를 포함합니다.\n\n예:\n\n```python\n# Automatically downloads blocks data if needed, then runs the SQL query\nclient.query_blockchain_sql(\n sql_query=\"SELECT block_number, gas_used, timestamp FROM blocks ORDER BY gas_used DESC LIMIT 10\",\n dataset=\"blocks\",\n blocks_from_latest=100\n)\n```\n\n### `list_available_sql_tables()`\n\nSQL로 쿼리할 수 있는 모든 사용 가능한 테이블을 나열하세요.\n\n예:\n\n```python\nclient.list_available_sql_tables()\n```\n\n### `get_sql_table_schema()`\n\n특정 Parquet 파일에 대한 스키마를 가져옵니다.\n\n매개변수:\n\n* `file_path` (str): Parquet 파일 경로\n\n예:\n\n```python\nclient.get_sql_table_schema(\"/path/to/blocks.parquet\")\n```\n\n### `get_sql_examples()`\n\n다양한 블록체인 데이터 세트에 대한 SQL 쿼리 예를 가져옵니다.\n\n예:\n\n```python\nclient.get_sql_examples()\n```\n\n## 구성 옵션\n\nCryo MCP 서버를 시작할 때 다음 명령줄 옵션을 사용할 수 있습니다.\n\n* `--rpc-url URL` : Ethereum RPC URL(ETH\\_RPC\\_URL 환경 변수보다 우선함)\n* `--data-dir PATH` : 다운로드한 데이터를 저장할 디렉토리(CRYO\\_DATA\\_DIR 환경 변수보다 우선하며, 기본값은 ~/.cryo-mcp/data/)\n\n## 환경 변수\n\n* `ETH_RPC_URL` : 명령줄을 통해 지정되지 않은 경우 사용할 기본 Ethereum RPC URL\n* `CRYO_DATA_DIR` : 명령줄을 통해 지정하지 않을 경우 다운로드된 데이터를 저장할 기본 디렉토리\n\n## 고급 사용법\n\n### 블록체인 데이터에 대한 SQL 쿼리\n\nCryo MCP를 사용하면 SQL의 유연성과 Cryo의 데이터 추출 기능을 결합하여 블록체인 데이터에 대해 강력한 SQL 쿼리를 실행할 수 있습니다.\n\n#### 2단계 SQL 쿼리 흐름\n\n데이터 추출과 쿼리를 두 개의 별도 단계로 나눌 수 있습니다.\n\n```python\n# Step 1: Download data and get file paths\ndownload_result = client.query_dataset(\n dataset=\"transactions\",\n blocks_from_latest=1000,\n output_format=\"parquet\"\n)\n\n# Step 2: Use the file paths to run SQL queries\nfile_paths = download_result.get(\"files\", [])\nclient.query_sql(\n query=f\"\"\"\n SELECT \n to_address as contract_address, \n COUNT(*) as tx_count,\n SUM(gas_used) as total_gas,\n AVG(gas_used) as avg_gas\n FROM read_parquet('{file_paths[0]}')\n WHERE to_address IS NOT NULL\n GROUP BY to_address\n ORDER BY total_gas DESC\n LIMIT 20\n \"\"\",\n files=file_paths\n)\n```\n\n#### 결합된 SQL 쿼리 흐름\n\n편의상 두 단계를 모두 처리하는 결합된 함수를 사용할 수도 있습니다.\n\n```python\n# Get top gas-consuming contracts\nclient.query_blockchain_sql(\n sql_query=\"\"\"\n SELECT \n to_address as contract_address, \n COUNT(*) as tx_count,\n SUM(gas_used) as total_gas,\n AVG(gas_used) as avg_gas\n FROM read_parquet('/path/to/transactions.parquet')\n WHERE to_address IS NOT NULL\n GROUP BY to_address\n ORDER BY total_gas DESC\n LIMIT 20\n \"\"\",\n dataset=\"transactions\",\n blocks_from_latest=1000\n)\n\n# Find blocks with the most transactions\nclient.query_blockchain_sql(\n sql_query=\"\"\"\n SELECT \n block_number, \n COUNT(*) as tx_count\n FROM read_parquet('/path/to/transactions.parquet')\n GROUP BY block_number\n ORDER BY tx_count DESC\n LIMIT 10\n \"\"\",\n dataset=\"transactions\",\n blocks=\"15M:16M\"\n)\n\n# Analyze event logs by topic\nclient.query_blockchain_sql(\n sql_query=\"\"\"\n SELECT \n topic0, \n COUNT(*) as event_count\n FROM read_parquet('/path/to/logs.parquet')\n GROUP BY topic0\n ORDER BY event_count DESC\n LIMIT 20\n \"\"\",\n dataset=\"logs\",\n blocks_from_latest=100\n)\n```\n\n**참고** : SQL 쿼리의 경우 DuckDB에서 최적의 성능을 보장하기 위해 데이터를 다운로드할 때 항상 `output_format=\"parquet\"` 사용하세요. `query_blockchain_sql` 사용하는 경우, `read_parquet()` 함수를 사용하여 SQL에서 파일 경로를 직접 참조해야 합니다.\n\n### 블록 범위를 사용한 쿼리\n\nCryo MCP는 Cryo의 블록 사양 구문의 전체 범위를 지원합니다.\n\n```python\n# Using block numbers\nclient.query_dataset('transactions', blocks='15000000:15001000')\n\n# Using K/M notation\nclient.query_dataset('logs', blocks='15M:15.01M')\n\n# Using offsets from latest \nclient.query_dataset('blocks', blocks_from_latest=100)\n```\n\n### 계약 필터링\n\n계약 주소로 로그 및 기타 데이터를 필터링합니다.\n\n```python\n# Get all logs for USDC contract\nclient.query_dataset('logs', \n blocks='16M:16.1M', \n contract='0xa0b86991c6218b36c1d19d4a2e9eb0ce3606eb48')\n```\n\n### 열 선택\n\n필요한 열만 포함하세요:\n\n```python\n# Get just block numbers and timestamps\nclient.query_dataset('blocks', \n blocks='16M:16.1M', \n include_columns=['number', 'timestamp'])\n```\n\n## 개발\n\n### 프로젝트 구조\n\n```\ncryo-mcp/\n├── cryo_mcp/ # Main package directory\n│ ├── __init__.py # Package initialization\n│ ├── server.py # Main MCP server implementation\n│ ├── sql.py # SQL query functionality\n├── tests/ # Test directory\n│ ├── test_*.py # Test files\n├── pyproject.toml # Project configuration\n├── README.md # Project documentation\n```\n\n### 테스트 실행\n\n`uv run pytest`\n\n## 특허\n\nMIT\n\n## 크레딧\n\n* Paradigm의 놀라운 [Cryo](https://github.com/paradigmxyz/cryo) 도구를 기반으로 구축됨\n* API 통신을 위해 [MCP 프로토콜을](https://github.com/mcp-team/mcp) 사용합니다.","Korean","ko-KR",{"_29":62,"_33":66,"_38":67},{"_31":63,"_33":64,"_35":65},"一个实现用于 Cryo 区块链数据提取的模型完成协议 (MCP) 的 API 服务器，允许用户通过任何兼容 MCP 的客户端查询以太坊区块链数据。","Cryo MCP 服务器","# 低温 MCP 🧊\n\n[Cryo](https://github.com/paradigmxyz/cryo)区块链数据提取工具的模型完成协议 (MCP) 服务器。\n\nCryo MCP 允许您通过实现 MCP 协议的 API 服务器访问 Cryo 强大的区块链数据提取功能，从而可以轻松地从任何兼容 MCP 的客户端查询区块链数据。\n\n## 对于 LLM 用户：SQL 查询工作流指南\n\n使用此 MCP 服务器对区块链数据运行 SQL 查询时，请遵循以下工作流程：\n\n1. 使用`query_dataset`**下载数据**：\n\n ```python\n result = query_dataset(\n dataset=\"blocks\", # or \"transactions\", \"logs\", etc.\n blocks=\"15000000:15001000\", # or use blocks_from_latest=100\n output_format=\"parquet\" # important: use parquet for SQL\n )\n files = result.get(\"files\", []) # Get the returned file paths\n ```\n\n2. 使用`get_sql_table_schema`**探索模式**：\n\n ```python\n # Check what columns are available in the file\n schema = get_sql_table_schema(files[0])\n # Now you can see all columns, data types, and sample data\n ```\n\n3. 使用`query_sql`**运行 SQL** ：\n\n ```python\n # Option 1: Simple table reference (DuckDB will match the table name to file)\n sql_result = query_sql(\n query=\"SELECT block_number, timestamp, gas_used FROM blocks\",\n files=files # Pass the files from step 1\n )\n\n # Option 2: Using read_parquet() with explicit file path\n sql_result = query_sql(\n query=f\"SELECT block_number, timestamp, gas_used FROM read_parquet('{files[0]}')\",\n files=files # Pass the files from step 1\n )\n ```\n\n或者，使用与`query_blockchain_sql`组合方法：\n\n```python\n# Option 1: Simple table reference\nresult = query_blockchain_sql(\n sql_query=\"SELECT * FROM blocks\",\n dataset=\"blocks\",\n blocks_from_latest=100\n)\n\n# Option 2: Using read_parquet()\nresult = query_blockchain_sql(\n sql_query=\"SELECT * FROM read_parquet('/path/to/file.parquet')\", # Path doesn't matter\n dataset=\"blocks\",\n blocks_from_latest=100\n)\n```\n\n有关完整的工作示例，请参阅[examples/sql\\_workflow\\_example.py](examples/sql_workflow_example.py) 。\n\n## 特征\n\n* **完整 Cryo 数据集访问**：通过 API 服务器查询任何 Cryo 数据集\n* **MCP 集成**：与 MCP 客户端无缝协作\n* **灵活的查询选项**：支持所有主要的Cryo过滤和输出选项\n* **区块范围选项**：查询特定区块、最新区块或相对范围\n* **合约过滤**：按合约地址过滤数据\n* **最新区块访问**：轻松访问最新的以太坊区块数据\n* **多种输出格式**：JSON、CSV 和 Parquet 支持\n* **架构信息**：获取详细的数据集架构和示例数据\n* **SQL 查询**：直接针对下载的区块链数据运行 SQL 查询\n\n## 安装（可选）\n\n如果您直接使用`uvx`运行该工具，则不需要这样做。\n\n```bash\n# install with UV (recommended)\nuv tool install cryo-mcp\n```\n\n## 要求\n\n* Python 3.8+\n* 紫外线\n* [Cryo](https://github.com/paradigmxyz/cryo)的工作安装\n* 访问以太坊 RPC 端点\n* DuckDB（用于 SQL 查询功能）\n\n## 快速入门\n\n### 与 Claude Code 一起使用\n\n1. 运行`claude mcp add`以获得交互式提示。\n2. 输入`uvx`作为运行命令。\n3. 输入`cryo-mcp --rpc-url [--data-dir ]`作为参数\n4. 或者，提供`ETH_RPC_URL`和`CRYO_DATA_DIR`作为环境变量。\n\n`claude`的新实例现在可以访问 cryo，因为它配置为命中您的 RPC 端点并将数据存储在指定的目录中。\n\n## 可用工具\n\nCryo MCP 公开了以下 MCP 工具：\n\n### `list_datasets()`\n\n返回所有可用 Cryo 数据集的列表。\n\n例子：\n\n```python\nclient.list_datasets()\n```\n\n### `query_dataset()`\n\n使用各种过滤选项查询 Cryo 数据集。\n\n参数：\n\n* `dataset` （str）：要查询的数据集的名称（例如，“blocks”、“transactions”、“logs”）\n* `blocks` （str，可选）：块范围规范（例如，'1000：1010'）\n* `start_block` (int，可选)：起始块号（块的替代）\n* `end_block` （int，可选）：结束块号（块的替代）\n* `use_latest` (bool, 可选): 如果为 True，则查询最新区块\n* `blocks_from_latest` (int，可选)：从最新到包含的块数\n* `contract` （str，可选）：要过滤的合约地址\n* `output_format` （str，可选）：输出格式（'json'，'csv'，'parquet'）\n* `include_columns` （列表，可选）：与默认值一起包含的列\n* `exclude_columns` （列表，可选）：从默认值中排除的列\n\n例子：\n\n```python\n# Get transactions from blocks 15M to 15.01M\nclient.query_dataset('transactions', blocks='15M:15.01M')\n\n# Get logs for a specific contract from the latest 100 blocks\nclient.query_dataset('logs', blocks_from_latest=100, contract='0x1234...')\n\n# Get just the latest block\nclient.query_dataset('blocks', use_latest=True)\n```\n\n### `lookup_dataset()`\n\n获取有关特定数据集的详细信息，包括模式和示例数据。\n\n参数：\n\n* `name` (str): 要查找的数据集的名称\n* `sample_start_block` （int，可选）：样本数据的起始块\n* `sample_end_block` （int，可选）：样本数据的结束块\n* `use_latest_sample` (bool，可选)：使用最新块作为样本\n* `sample_blocks_from_latest` (int，可选)：样本的最新块数\n\n例子：\n\n```python\nclient.lookup_dataset('logs')\n```\n\n### `get_latest_ethereum_block()`\n\n返回有关最新以太坊区块的信息。\n\n例子：\n\n```python\nclient.get_latest_ethereum_block()\n```\n\n### SQL查询工具\n\nCryo MCP 包含几个用于针对区块链数据运行 SQL 查询的工具：\n\n### `query_sql()`\n\n对下载的区块链数据运行 SQL 查询。\n\n参数：\n\n* `query` （str）：要执行的 SQL 查询\n* `files` （列表，可选）：要查询的 parquet 文件路径列表。如果为 None，则使用数据目录中的所有文件。\n* `include_schema` (bool，可选)：是否在结果中包含架构信息\n\n例子：\n\n```python\n# Run against all available files\nclient.query_sql(\"SELECT * FROM read_parquet('/path/to/blocks.parquet') LIMIT 10\")\n\n# Run against specific files\nclient.query_sql(\n \"SELECT * FROM read_parquet('/path/to/blocks.parquet') LIMIT 10\",\n files=['/path/to/blocks.parquet']\n)\n```\n\n### `query_blockchain_sql()`\n\n使用 SQL 查询区块链数据，自动下载任何所需数据。\n\n参数：\n\n* `sql_query` (str): 要执行的 SQL 查询\n* `dataset` （str，可选）：要查询的数据集（例如，“区块”、“交易”）\n* `blocks` （str，可选）：块范围规范\n* `start_block` （int，可选）：起始块号\n* `end_block` （int，可选）：结束块号\n* `use_latest` (bool, 可选): 如果为 True，则查询最新区块\n* `blocks_from_latest` (int，可选)：要包含的最新块之前块的数量\n* `contract` （str，可选）：要过滤的合约地址\n* `force_refresh` （bool，可选）：即使存在新数据也强制下载。\n* `include_schema` (bool，可选)：在结果中包含架构信息\n\n例子：\n\n```python\n# Automatically downloads blocks data if needed, then runs the SQL query\nclient.query_blockchain_sql(\n sql_query=\"SELECT block_number, gas_used, timestamp FROM blocks ORDER BY gas_used DESC LIMIT 10\",\n dataset=\"blocks\",\n blocks_from_latest=100\n)\n```\n\n### `list_available_sql_tables()`\n\n列出所有可用 SQL 查询的表。\n\n例子：\n\n```python\nclient.list_available_sql_tables()\n```\n\n### `get_sql_table_schema()`\n\n获取特定 parquet 文件的架构。\n\n参数：\n\n* `file_path` (str): parquet 文件的路径\n\n例子：\n\n```python\nclient.get_sql_table_schema(\"/path/to/blocks.parquet\")\n```\n\n### `get_sql_examples()`\n\n获取不同区块链数据集的示例 SQL 查询。\n\n例子：\n\n```python\nclient.get_sql_examples()\n```\n\n## 配置选项\n\n启动 Cryo MCP 服务器时，您可以使用以下命令行选项：\n\n* `--rpc-url URL` ：以太坊 RPC URL（覆盖 ETH\\_RPC\\_URL 环境变量）\n* `--data-dir PATH` ：存储下载数据的目录（覆盖 CRYO\\_DATA\\_DIR 环境变量，默认为 ~/.cryo-mcp/data/）\n\n## 环境变量\n\n* `ETH_RPC_URL` ：未通过命令行指定时使用的默认以太坊 RPC URL\n* `CRYO_DATA_DIR` ：未通过命令行指定时存储下载数据的默认目录\n\n## 高级用法\n\n### 针对区块链数据的 SQL 查询\n\nCryo MCP 允许您对区块链数据运行强大的 SQL 查询，将 SQL 的灵活性与 Cryo 的数据提取功能相结合：\n\n#### 两步 SQL 查询流程\n\n您可以将数据提取和查询分为两个独立的步骤：\n\n```python\n# Step 1: Download data and get file paths\ndownload_result = client.query_dataset(\n dataset=\"transactions\",\n blocks_from_latest=1000,\n output_format=\"parquet\"\n)\n\n# Step 2: Use the file paths to run SQL queries\nfile_paths = download_result.get(\"files\", [])\nclient.query_sql(\n query=f\"\"\"\n SELECT \n to_address as contract_address, \n COUNT(*) as tx_count,\n SUM(gas_used) as total_gas,\n AVG(gas_used) as avg_gas\n FROM read_parquet('{file_paths[0]}')\n WHERE to_address IS NOT NULL\n GROUP BY to_address\n ORDER BY total_gas DESC\n LIMIT 20\n \"\"\",\n files=file_paths\n)\n```\n\n#### 组合 SQL 查询流\n\n为了方便起见，您还可以使用处理这两个步骤的组合函数：\n\n```python\n# Get top gas-consuming contracts\nclient.query_blockchain_sql(\n sql_query=\"\"\"\n SELECT \n to_address as contract_address, \n COUNT(*) as tx_count,\n SUM(gas_used) as total_gas,\n AVG(gas_used) as avg_gas\n FROM read_parquet('/path/to/transactions.parquet')\n WHERE to_address IS NOT NULL\n GROUP BY to_address\n ORDER BY total_gas DESC\n LIMIT 20\n \"\"\",\n dataset=\"transactions\",\n blocks_from_latest=1000\n)\n\n# Find blocks with the most transactions\nclient.query_blockchain_sql(\n sql_query=\"\"\"\n SELECT \n block_number, \n COUNT(*) as tx_count\n FROM read_parquet('/path/to/transactions.parquet')\n GROUP BY block_number\n ORDER BY tx_count DESC\n LIMIT 10\n \"\"\",\n dataset=\"transactions\",\n blocks=\"15M:16M\"\n)\n\n# Analyze event logs by topic\nclient.query_blockchain_sql(\n sql_query=\"\"\"\n SELECT \n topic0, \n COUNT(*) as event_count\n FROM read_parquet('/path/to/logs.parquet')\n GROUP BY topic0\n ORDER BY event_count DESC\n LIMIT 20\n \"\"\",\n dataset=\"logs\",\n blocks_from_latest=100\n)\n```\n\n**注意**：对于 SQL 查询，下载数据时请始终使用`output_format=\"parquet\"`以确保 DuckDB 获得最佳性能。使用`query_blockchain_sql`时，应使用`read_parquet()`函数直接在 SQL 中引用文件路径。\n\n### 使用块范围查询\n\nCryo MCP 支持 Cryo 的全部块规范语法：\n\n```python\n# Using block numbers\nclient.query_dataset('transactions', blocks='15000000:15001000')\n\n# Using K/M notation\nclient.query_dataset('logs', blocks='15M:15.01M')\n\n# Using offsets from latest \nclient.query_dataset('blocks', blocks_from_latest=100)\n```\n\n### 合同过滤\n\n按合约地址过滤日志和其他数据：\n\n```python\n# Get all logs for USDC contract\nclient.query_dataset('logs', \n blocks='16M:16.1M', \n contract='0xa0b86991c6218b36c1d19d4a2e9eb0ce3606eb48')\n```\n\n### 列选择\n\n仅包含您需要的列：\n\n```python\n# Get just block numbers and timestamps\nclient.query_dataset('blocks', \n blocks='16M:16.1M', \n include_columns=['number', 'timestamp'])\n```\n\n## 发展\n\n### 项目结构\n\n```\ncryo-mcp/\n├── cryo_mcp/ # Main package directory\n│ ├── __init__.py # Package initialization\n│ ├── server.py # Main MCP server implementation\n│ ├── sql.py # SQL query functionality\n├── tests/ # Test directory\n│ ├── test_*.py # Test files\n├── pyproject.toml # Project configuration\n├── README.md # Project documentation\n```\n\n### 运行测试\n\n`uv run pytest`\n\n## 执照\n\n麻省理工学院\n\n## 致谢\n\n* 基于 Paradigm 出色的[Cryo](https://github.com/paradigmxyz/cryo)工具构建\n* 使用[MCP 协议](https://github.com/mcp-team/mcp)进行 API 通信","Chinese","zh-CN","githubUser","locale",{"_71":72,"_73":74,"_13":85,"_95":32,"_96":32,"_97":98,"_101":102,"_125":-5,"_16":126,"_127":128,"_129":130,"_33":34,"_148":115,"_149":-5,"_150":-5,"_151":152,"_153":-5,"_154":155,"_160":-5,"_137":111,"_161":162,"_166":167,"_116":291,"_292":293},"addedAt",["D",1741489309785],"argumentsJsonSchema",{"_75":76,"_82":83,"_80":84},"properties",{"_77":78},"ETH_RPC_URL",{"_31":79,"_80":81},"Default Ethereum RPC URL to use when not specified via command line","type","string","required",[77],"object",[86,87,88,89,90,91,92,93,94],"category:blockchain","category:databases","category:web3","environment:network","hosting:hybrid","language:python","security_grade:a","quality_grade:a","license_grade:a","descriptionMarkdown","descriptionPlainText","githubProject",{"_35":36,"_99":100},"url","https://github.com/z80dev/cryo-mcp","githubRepository",{"_103":104,"_105":106,"_107":108,"_109":110,"_33":111,"_112":113,"_118":119,"_120":121,"_123":124},"defaultBranch","main","fullName","z80dev/cryo-mcp","githubTreeId",77976,"language","Python","cryo-mcp","owner",{"_114":115,"_116":117},"login","z80dev","uid","U1colCoU5z","recentStargazers",6,"spdxLicense",{"_33":122},"MIT License","stargazers",86,"glamaJson",1728,"inspectable",true,"integrations",[131,141],{"_132":133,"_31":139,"_116":140},"brand",{"_134":135,"_33":136,"_137":138},"logoSvgPath","M12 0C5.363 0 0 5.363 0 12s5.363 12 12 12 12-5.363 12-12S18.637 0 12 0zM9.502 7.03a4.974 4.974 0 0 1 4.97 4.97 4.974 4.974 0 0 1-4.97 4.97A4.974 4.974 0 0 1 4.532 12a4.974 4.974 0 0 1 4.97-4.97zm6.563 3.183h2.351c.98 0 1.787.782 1.787 1.762s-.807 1.789-1.787 1.789h-2.351v-3.551z","DuckDB","slug","duckdb","Enables SQL queries against blockchain data through DuckDB, supporting data exploration, schema inspection, and complex analytics on downloaded blockchain datasets with various filtering and aggregation capabilities.","j9yw7zquie",{"_132":142,"_31":146,"_116":147},{"_134":143,"_33":144,"_137":145},"M11.944 17.97L4.58 13.62 11.943 24l7.37-10.38-7.372 4.35h.003zM12.056 0L4.69 12.223l7.365 4.354 7.365-4.35L12.056 0z","Ethereum","ethereum","Provides blockchain data extraction capabilities via an API server, allowing querying of block, transaction, and log data from Ethereum with support for block range specifications, contract filtering, and multiple output formats.","0jp2oqv3x1","namespace","npmPackage","prompts","redditPostForeignId","1jb1gru","resources","scores",{"_156":157,"_158":157,"_159":157},"license",100,"quality","security","simpleIcon","supportedPlatforms",[163,164,165],"WINDOWS","LINUX","MACOS","tools",[168,176,245,275,285],{"_33":169,"_31":170,"_171":172},"list_datasets","Return a list of all available cryo datasets","inputSchema",{"_80":84,"_173":174,"_75":175},"title","list_datasetsArguments",{},{"_33":177,"_31":178,"_171":179},"query_dataset","\n Query a cryo dataset and return the results\n\n Args:\n dataset: The name of the dataset to query (e.g., 'logs', 'transactions')\n blocks: Block range specification as a string (e.g., '1000:1010')\n start_block: Start block number as integer (alternative to blocks)\n end_block: End block number as integer (alternative to blocks)\n use_latest: If True, query the latest block\n blocks_from_latest: Number of blocks before the latest to include (e.g., 10 = latest-10 to latest)\n contract: Contract address to filter by\n output_format: Output format (json, csv, parquet)\n include_columns: Columns to include alongside the defaults\n exclude_columns: Columns to exclude from the defaults\n\n Returns:\n The dataset results\n ",{"_80":84,"_173":180,"_82":181,"_75":183},"query_datasetArguments",[182],"dataset",{"_184":185,"_182":193,"_195":196,"_201":202,"_208":209,"_213":214,"_219":220,"_223":224,"_232":233,"_239":240},"blocks",{"_186":187,"_173":191,"_192":-5},"anyOf",[188,189],{"_80":81},{"_80":190},"null","Blocks","default",{"_80":81,"_173":194},"Dataset","contract",{"_186":197,"_173":200,"_192":-5},[198,199],{"_80":81},{"_80":190},"Contract","end_block",{"_186":203,"_173":207,"_192":-5},[204,206],{"_80":205},"integer",{"_80":190},"End Block","use_latest",{"_80":210,"_173":211,"_192":212},"boolean","Use Latest",false,"start_block",{"_186":215,"_173":218,"_192":-5},[216,217],{"_80":205},{"_80":190},"Start Block","output_format",{"_80":81,"_173":221,"_192":222},"Output Format","json","exclude_columns",{"_186":225,"_173":231,"_192":-5},[226,230],{"_80":227,"_228":229},"array","items",{"_80":81},{"_80":190},"Exclude Columns","include_columns",{"_186":234,"_173":238,"_192":-5},[235,237],{"_80":227,"_228":236},{"_80":81},{"_80":190},"Include Columns","blocks_from_latest",{"_186":241,"_173":244,"_192":-5},[242,243],{"_80":205},{"_80":190},"Blocks From Latest",{"_33":246,"_31":247,"_171":248},"lookup_dataset","\n Look up a specific dataset and return detailed information about it\n \n Args:\n name: The name of the dataset to look up\n sample_start_block: Optional start block for sample data (integer)\n sample_end_block: Optional end block for sample data (integer)\n use_latest_sample: If True, use the latest block for sample data\n sample_blocks_from_latest: Number of blocks before the latest to include in sample\n \n Returns:\n Detailed information about the dataset including schema and available fields\n ",{"_80":84,"_173":249,"_82":250,"_75":251},"lookup_datasetArguments",[33],{"_33":252,"_254":255,"_260":261,"_263":264,"_269":270},{"_80":81,"_173":253},"Name","sample_end_block",{"_186":256,"_173":259,"_192":-5},[257,258],{"_80":205},{"_80":190},"Sample End Block","use_latest_sample",{"_80":210,"_173":262,"_192":212},"Use Latest Sample","sample_start_block",{"_186":265,"_173":268,"_192":-5},[266,267],{"_80":205},{"_80":190},"Sample Start Block","sample_blocks_from_latest",{"_186":271,"_173":274,"_192":-5},[272,273],{"_80":205},{"_80":190},"Sample Blocks From Latest",{"_33":276,"_31":277,"_171":278},"get_transaction_by_hash","\n Get detailed information about a transaction by its hash\n \n Args:\n tx_hash: The transaction hash to look up\n \n Returns:\n Detailed information about the transaction\n ",{"_80":84,"_173":279,"_82":280,"_75":282},"get_transaction_by_hashArguments",[281],"tx_hash",{"_281":283},{"_80":81,"_173":284},"Tx Hash",{"_33":286,"_31":287,"_171":288},"get_latest_ethereum_block","\n Get information about the latest Ethereum block\n \n Returns:\n Information about the latest block including block number\n ",{"_80":84,"_173":289,"_75":290},"get_latest_ethereum_blockArguments",{},"90ftd26na5","updatedAt",["D",1742491686295],"meta",[296,298,300,307,310,312,315,318,321,324,327,329,332,335,337,339,341,343,345],{"_173":297},"Cryo MCP Server | Glama",{"_299":32,"_33":31},"content",{"_301":302,"_303":304,"_305":306},"href","https://glama.ai/mcp/servers/@z80dev/cryo-mcp","rel","canonical","tagName","link",{"_299":34,"_308":309},"property","og:title",{"_299":32,"_308":311},"og:description",{"_299":313,"_308":314},"https://glama.ai/logo.png","og:logo",{"_299":316,"_308":317},"https://glama.ai/mcp/servers/@z80dev/cryo-mcp/og-image","og:image",{"_299":319,"_308":320},"630","og:image:height",{"_299":322,"_308":323},"1200","og:image:width",{"_299":325,"_308":326},"website","og:type",{"_299":302,"_308":328},"og:url",{"_299":330,"_308":331},"Glama – MCP Hosting Platform","og:site_name",{"_299":333,"_33":334},"summary_large_image","twitter:card",{"_299":330,"_33":336},"twitter:title",{"_299":32,"_33":338},"twitter:description",{"_299":316,"_33":340},"twitter:image",{"_299":319,"_33":342},"twitter:image:height",{"_299":322,"_33":344},"twitter:image:width",{"_346":347},"script:ld+json",{"_348":349,"_350":351},"@context","http://schema.org","@graph",[352],{"_348":353,"_354":355,"_356":357},"https://schema.org","@type","BreadcrumbList","itemListElement",[358,365,369,373],{"_354":359,"_360":361,"_33":362,"_363":364},"ListItem","item","https://glama.ai/","Glama","position",1,{"_354":359,"_360":366,"_33":367,"_363":368},"https://glama.ai/mcp","MCP",2,{"_354":359,"_360":370,"_33":371,"_363":372},"https://glama.ai/mcp/servers","Servers",3,{"_354":359,"_360":302,"_33":34,"_363":374},4,"visitorIsMaintainer","routes/_public/mcp/servers/~namespace/~slug/_pages/_explorer/blob/~branch/~/_route",{"_378":379,"_383":104,"_29":384,"_294":470,"_494":495,"_496":497},"blob",{"_299":380,"_381":382},"\"\"\"\nCryo MCP Server - Query Ethereum blockchain data using cryo and MCP\n\"\"\"\n\n__version__ = \"0.1.4\"\n","fileName","__init__.py","branch",{"_71":385,"_73":386,"_13":390,"_95":32,"_96":32,"_97":391,"_101":392,"_125":-5,"_16":126,"_127":128,"_129":395,"_33":34,"_148":115,"_149":-5,"_150":-5,"_151":152,"_153":-5,"_154":400,"_160":-5,"_137":111,"_161":401,"_166":402,"_116":291,"_292":469},["D",1741489309785],{"_75":387,"_82":389,"_80":84},{"_77":388},{"_31":79,"_80":81},[77],[86,87,88,89,90,91,92,93,94],{"_35":36,"_99":100},{"_103":104,"_105":106,"_107":108,"_109":110,"_33":111,"_112":393,"_118":119,"_120":394,"_123":124},{"_114":115,"_116":117},{"_33":122},[396,398],{"_132":397,"_31":139,"_116":140},{"_134":135,"_33":136,"_137":138},{"_132":399,"_31":146,"_116":147},{"_134":143,"_33":144,"_137":145},{"_156":157,"_158":157,"_159":157},[163,164,165],[403,406,443,461,466],{"_33":169,"_31":170,"_171":404},{"_80":84,"_173":174,"_75":405},{},{"_33":177,"_31":178,"_171":407},{"_80":84,"_173":180,"_82":408,"_75":409},[182],{"_184":410,"_182":414,"_195":415,"_201":419,"_208":423,"_213":424,"_219":428,"_223":429,"_232":434,"_239":439},{"_186":411,"_173":191,"_192":-5},[412,413],{"_80":81},{"_80":190},{"_80":81,"_173":194},{"_186":416,"_173":200,"_192":-5},[417,418],{"_80":81},{"_80":190},{"_186":420,"_173":207,"_192":-5},[421,422],{"_80":205},{"_80":190},{"_80":210,"_173":211,"_192":212},{"_186":425,"_173":218,"_192":-5},[426,427],{"_80":205},{"_80":190},{"_80":81,"_173":221,"_192":222},{"_186":430,"_173":231,"_192":-5},[431,433],{"_80":227,"_228":432},{"_80":81},{"_80":190},{"_186":435,"_173":238,"_192":-5},[436,438],{"_80":227,"_228":437},{"_80":81},{"_80":190},{"_186":440,"_173":244,"_192":-5},[441,442],{"_80":205},{"_80":190},{"_33":246,"_31":247,"_171":444},{"_80":84,"_173":249,"_82":445,"_75":446},[33],{"_33":447,"_254":448,"_260":452,"_263":453,"_269":457},{"_80":81,"_173":253},{"_186":449,"_173":259,"_192":-5},[450,451],{"_80":205},{"_80":190},{"_80":210,"_173":262,"_192":212},{"_186":454,"_173":268,"_192":-5},[455,456],{"_80":205},{"_80":190},{"_186":458,"_173":274,"_192":-5},[459,460],{"_80":205},{"_80":190},{"_33":276,"_31":277,"_171":462},{"_80":84,"_173":279,"_82":463,"_75":464},[281],{"_281":465},{"_80":81,"_173":284},{"_33":286,"_31":287,"_171":467},{"_80":84,"_173":289,"_75":468},{},["D",1742491686295],[471,473,475,477,479,480,481,483,484,485,486,487,488,489,490,491,492,493],{"_173":472},"MCP Server | Glama",{"_299":474,"_33":31},"Enterprise-grade security, privacy, and features like agents, MCP, prompt templates, and more.",{"_301":476,"_303":304,"_305":306},"https://glama.ai/mcp/servers/@z80dev/cryo-mcp/blob/main/cryo_mcp",{"_299":478,"_308":309},"MCP Server",{"_299":474,"_308":311},{"_299":313,"_308":314},{"_299":482,"_308":317},"https://glama.ai/generated-images/og?title=MCP+Server",{"_299":319,"_308":320},{"_299":322,"_308":323},{"_299":325,"_308":326},{"_299":476,"_308":328},{"_299":330,"_308":331},{"_299":333,"_33":334},{"_299":330,"_33":336},{"_299":474,"_33":338},{"_299":482,"_33":340},{"_299":319,"_33":342},{"_299":322,"_33":344},"path","cryo_mcp","treeMembers",[498,500,502],{"_494":499,"_80":378},"cryo_mcp/__init__.py",{"_494":501,"_80":378},"cryo_mcp/server.py",{"_494":503,"_80":378},"cryo_mcp/sql.py","actionData","errors"]