MCP Trino Server
MCP Trino 服务器
MCP Trino 服务器是一个模型上下文协议 (MCP)服务器,可与 Trino 和 Iceberg 无缝集成,通过标准接口实现高级数据探索、查询和表维护功能。
用例
Trino 中的交互式数据探索和分析
自动化 Iceberg 表维护和优化
构建与 Trino 数据库交互的人工智能工具
使用适当的结果格式执行和管理 SQL 查询
Related MCP server: Alibaba Cloud DMS MCP Server
先决条件
正在运行的 Trino 服务器(或用于本地开发的 Docker Compose)
Python 3.11 或更高版本
Docker(可选,用于容器化部署)
安装
通过 Smithery 安装
要通过Smithery自动为 Claude Desktop 安装 MCP Trino Server:
npx -y @smithery/cli install @alaturqua/mcp-trino-python --client claude本地运行 Trino
最简单的入门方法是使用包含的 Docker Compose 配置在本地运行 Trino:
docker-compose up -d这将在localhost:8080上启动 Trino 服务器。现在您可以继续配置 MCP 服务器。
与 VS Code 一起使用
为了快速安装,您可以将以下配置添加到 VS Code 设置中。您可以按Ctrl + Shift + P并输入Preferences: Open User Settings (JSON)来完成此操作。
或者,您可以将其添加到工作区中名为.vscode/mcp.json的文件中。这样您就可以与其他人共享该配置。
请注意
.vscode/mcp.json文件中不需要mcp键。
{
"mcp": {
"servers": {
"trino": {
"command": "docker",
"args": ["run", "--rm", "ghcr.io/alaturqua/mcp-trino-python:latest"],
"env": {
"TRINO_HOST": "${input:trino_host}",
"TRINO_PORT": "${input:trino_port}",
"TRINO_USER": "${input:trino_user}",
"TRINO_PASSWORD": "${input:trino_password}",
"TRINO_HTTP_SCHEME": "${input:trino_http_scheme}",
"TRINO_CATALOG": "${input:trino_catalog}",
"TRINO_SCHEMA": "${input:trino_schema}"
}
}
}
}
}与 Claude Desktop 一起使用
将以下配置添加到您的 Claude Desktop 设置:
{
"mcpServers": {
"trino": {
"command": "python",
"args": ["./src/server.py"],
"env": {
"TRINO_HOST": "your-trino-host",
"TRINO_PORT": "8080",
"TRINO_USER": "trino"
}
}
}
}配置
环境变量
多变的 | 描述 | 默认 |
TRINO_HOST | Trino 服务器主机名 | 本地主机 |
TRINO_PORT | Trino 服务器端口 | 8080 |
TRINO_用户 | Trino 用户名 | 特里诺 |
TRINO_目录 | 默认目录 | 没有任何 |
TRINO_SCHEMA | 默认架构 | 没有任何 |
TRINO_HTTP_方案 | HTTP 方案 (http/https) | http |
TRINO_密码 | Trino 密码 | 没有任何 |
工具
查询和探索工具
显示目录
列出所有可用目录
无需参数
显示模式
列出目录中的所有架构
参数:
catalog:目录名称(字符串,必需)
显示表
列出模式中的所有表
参数:
catalog:目录名称(字符串,必需)schema:架构名称(字符串,必需)
描述表
显示详细的表结构和列信息
参数:
table:表名(字符串,必需)catalog:目录名称(字符串,可选)schema:架构名称(字符串,可选)
执行查询
执行 SQL 查询并返回格式化的结果
参数:
query:要执行的 SQL 查询(字符串,必需)
显示目录树
显示目录、模式和表的层次树视图
返回带有视觉指示器的格式化树结构
无需参数
显示_创建_表
显示表的 CREATE TABLE 语句
参数:
table:表名(字符串,必需)catalog:目录名称(字符串,可选)schema:架构名称(字符串,可选)
显示_创建_视图
显示视图的 CREATE VIEW 语句
参数:
view:视图名称(字符串,必需)catalog:目录名称(字符串,可选)schema:架构名称(字符串,可选)
显示统计信息
显示表的统计数据
参数:
table:表名(字符串,必需)catalog:目录名称(字符串,可选)schema:架构名称(字符串,可选)
冰山桌维护
优化
通过压缩小文件来优化冰山表
参数:
table:表名(字符串,必需)catalog:目录名称(字符串,可选)schema:架构名称(字符串,可选)
优化清单
优化 Iceberg 表的清单文件
参数:
table:表名(字符串,必需)catalog:目录名称(字符串,可选)schema:架构名称(字符串,可选)
过期快照
从 Iceberg 表中删除旧快照
参数:
table:表名(字符串,必需)retention_threshold:年龄阈值(例如“7d”)(字符串,可选)catalog:目录名称(字符串,可选)schema:架构名称(字符串,可选)
Iceberg 元数据检查
显示表属性
显示 Iceberg 表属性
参数:
table:表名(字符串,必需)catalog:目录名称(字符串,可选)schema:架构名称(字符串,可选)
显示表历史记录
显示 Iceberg 表历史记录/更新日志
包含快照时间、谱系和祖先信息
参数:
table:表名(字符串,必需)catalog:目录名称(字符串,可选)schema:架构名称(字符串,可选)
显示元数据日志条目
显示 Iceberg 表元数据日志条目
包含元数据文件位置和序列信息
参数:
table:表名(字符串,必需)catalog:目录名称(字符串,可选)schema:架构名称(字符串,可选)
显示快照
显示 Iceberg 表快照
包含快照详细信息,包括操作和清单文件
参数:
table:表名(字符串,必需)catalog:目录名称(字符串,可选)schema:架构名称(字符串,可选)
显示清单
显示当前或所有快照的 Iceberg 表清单
包含清单文件详细信息和数据文件统计信息
参数:
table:表名(字符串,必需)catalog:目录名称(字符串,可选)schema:架构名称(字符串,可选)all_snapshots:包含所有快照(布尔值,可选)
显示分区
显示 Iceberg 表分区
包含分区统计信息和文件计数
参数:
table:表名(字符串,必需)catalog:目录名称(字符串,可选)schema:架构名称(字符串,可选)
显示文件
显示当前快照中的 Iceberg 表数据文件
包含详细的文件元数据和列统计信息
参数:
table:表名(字符串,必需)catalog:目录名称(字符串,可选)schema:架构名称(字符串,可选)
显示条目
显示当前或所有快照的 Iceberg 表清单条目
包含条目状态和详细文件指标
参数:
table:表名(字符串,必需)catalog:目录名称(字符串,可选)schema:架构名称(字符串,可选)all_snapshots:包含所有快照(布尔值,可选)
显示参考
显示 Iceberg 表引用(分支和标签)
包含参考配置和快照映射
参数:
table:表名(字符串,必需)catalog:目录名称(字符串,可选)schema:架构名称(字符串,可选)
查询历史记录
显示查询历史记录
获取已执行查询的历史记录
参数:
limit:返回的最大查询数(数字,可选)
执照
本项目遵循 Apache 2.0 许可证。请参阅许可证文件以了解完整条款。
Available Tools
22 toolsdescribe_tableC
Describe a table
Inputs: catalog (required), schema_name (required), table (required). Returns: result
execute_queryC
Execute a SQL query and return results in a readable format
Inputs: query (required). Returns: result
expire_snapshotsB
Remove old snapshots from an Iceberg table
Inputs: catalog (required), schema_name (required), table (required), retention_threshold. Returns: result
optimizeB
Optimize an Iceberg table's data files
Inputs: catalog (required), schema_name (required), table (required). Returns: result
optimize_manifestsC
Optimize manifest files for an Iceberg table
Inputs: catalog (required), schema_name (required), table (required). Returns: result
show_catalogsB
List all available catalogs
Inputs: no named parameters. Returns: result
show_catalog_treeA
Show a hierarchical tree view of catalogs, schemas, and tables
Inputs: no named parameters. Returns: result
show_create_tableA
Show the CREATE TABLE statement for a specific table
Inputs: catalog (required), schema_name (required), table (required). Returns: result
show_create_viewB
Show the CREATE VIEW statement for a specific view
Inputs: catalog (required), schema_name (required), view (required). Returns: result
show_entriesC
Show Iceberg table manifest entries
Inputs: catalog (required), schema_name (required), table (required), all_snapshots. Returns: result
show_filesC
Show Iceberg table data files
Inputs: catalog (required), schema_name (required), table (required). Returns: result
show_manifestsC
Show Iceberg table manifests
Inputs: catalog (required), schema_name (required), table (required), all_snapshots. Returns: result
show_metadata_log_entriesC
Show metadata for the table
Inputs: catalog (required), schema_name (required), table (required). Returns: result
show_partitionsB
Show Iceberg table partitions
Inputs: catalog (required), schema_name (required), table (required). Returns: result
show_query_historyB
Get the history of executed queries
Inputs: limit. Returns: result
show_refsA
Show Iceberg table references (branches and tags)
Inputs: catalog (required), schema_name (required), table (required). Returns: result
show_schemasB
List all schemas in a catalog
Inputs: catalog (required). Returns: result
show_snapshotsC
Show Iceberg table snapshots
Inputs: catalog (required), schema_name (required), table (required). Returns: result
show_statsC
Show statistics for a table
Inputs: catalog (required), schema_name (required), table (required). Returns: result
show_table_historyC
Show Iceberg table history/changelog
Inputs: catalog (required), schema_name (required), table (required). Returns: result
show_table_propertiesC
Show Iceberg table properties
Inputs: catalog (required), schema_name (required), table (required). Returns: result
show_tablesB
List all tables in a schema
Inputs: catalog (required), schema_name (required). Returns: result
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 22 tools
Most tools have distinct purposes, especially the 'show_' variants which target different Iceberg metadata. However, some concepts like show_table_history, show_metadata_log_entries, and show_snapshots overlap conceptually, causing minor potential confusion.
All tools follow a verb_noun pattern with underscores, which is consistent. The verbs vary (show, describe, execute, expire, optimize) but are clear and appropriate, though a fully uniform verb set would improve consistency.
22 tools is on the higher side but remains reasonable given the dual focus on general Trino queries and detailed Iceberg management. The scope justifies the count, though it could be split into two servers for clarity.
The tool set covers core query operations and comprehensive Iceberg maintenance (snapshots, files, optimization, history). Missing DDL commands like create/drop table or schema are notable gaps but align with the server's apparent administrative/analytical focus.
Maintenance
Related MCP Connectors
The BigQuery remote MCP server is a fully managed service that uses the Model Context Protocol to connect AI applications and LLMs to BigQuery data sources. It provides secure, standardized tools for AI agents to list datasets and tables, retrieve schemas, generate and execute SQL queries through natural language, and analyze data—enabling direct access to enterprise analytics data without requiring manual SQL coding.
A comprehensive Model Context Protocol (MCP) server that enables AI assistants to interact with yo…
A Model Context Protocol server for Wix AI tools
The Mercado Pago MCP Server implements the Model Context Protocol to provide AI agents and LLMs with access to Mercado Pago's APIs and tools within compatible development environments. It acts as an intermediary that translates Mercado Pago resources into executable functions (tools) that AI applications can invoke to perform actions and automate flows. The server simplifies integration, enables using documentation to implement or improve code, and optimizes operations through natural language interactions without manual implementations.
Related MCP Servers
- FlicenseBqualityDmaintenanceA Model Context Protocol server that provides a SQL interface for querying and managing Apache Iceberg tables through Claude desktop, allowing natural language interaction with Iceberg data lakes.18-
- AlicenseAqualityCmaintenanceA Model Context Protocol server that enables large language models to access database metadata and perform cross-engine data querying across diverse database ecosystems.1654Apache 2.0
- AlicenseNot gradedqualityDmaintenanceMCP server providing seamless integration with Trino and Iceberg for advanced data exploration, querying, and table maintenance.Apache 2.0
- AlicenseBqualityDmaintenanceModel Context Protocol server for read-only access to Iceberg tables through Cloudera Trino: schema discovery, SQL queries, metadata-based health checks, time travel, and performance analysis.173Apache 2.0