如何使用 Claude Agent SDK 和 ClickHouse MCP 服务器构建 AI Agent

1

安装库

运行以下命令安装 Claude Agent SDK 库：

pip install -q --upgrade pip
pip install -q claude-agent-sdk
pip install -q ipywidgets

2

设置凭据

接下来，你需要提供 Anthropic API key：

import os, getpass
os.environ["ANTHROPIC_API_KEY"] = getpass.getpass("Enter Anthropic API Key:")

响应

Enter Anthropic API Key: ········

接下来，定义连接到 ClickHouse SQL playground 所需的凭据：

env = {
    "CLICKHOUSE_HOST": "sql-clickhouse.clickhouse.com",
    "CLICKHOUSE_PORT": "8443",
    "CLICKHOUSE_USER": "demo",
    "CLICKHOUSE_PASSWORD": "",
    "CLICKHOUSE_SECURE": "true"
}

3

初始化 MCP 服务器和 Claude Agent SDK agent

现在将 ClickHouse MCP 服务器配置为指向 ClickHouse SQL playground，然后初始化 agent 并向它提一个问题：

from claude_agent_sdk import query, ClaudeAgentOptions, AssistantMessage, UserMessage, TextBlock, ToolUseBlock

options = ClaudeAgentOptions(
    allowed_tools=[
        "mcp__mcp-clickhouse__list_databases",
        "mcp__mcp-clickhouse__list_tables", 
        "mcp__mcp-clickhouse__run_select_query",
        "mcp__mcp-clickhouse__run_chdb_select_query"
    ],
    mcp_servers={
        "mcp-clickhouse": {
            "command": "uv",
            "args": [
                "run",
                "--with", "mcp-clickhouse",
                "--python", "3.10",
                "mcp-clickhouse"
            ],
            "env": env
        }
    }
)

async for message in query(prompt="Tell me something interesting about UK property sales", options=options):
    if isinstance(message, AssistantMessage):
        for block in message.content:
            if isinstance(block, TextBlock):
                print(f"🤖 {block.text}")
            if isinstance(block, ToolUseBlock):
                print(f"🛠️ {block.name} {block.input}")
    elif isinstance(message, UserMessage):
        for block in message.content:
            if isinstance(block, TextBlock):
                print(block.text)

请注意，for 代码块中的代码为了简洁起见对输出做了过滤。

响应

🤖 我会查询 ClickHouse 数据库，找一些关于英国房产交易的有趣信息。

先让我看看有哪些数据库可用：
🛠️ mcp__mcp-clickhouse__list_databases {}
🤖 很好！这里有一个 “uk” 数据库。让我看看有哪些表可用：
🛠️ mcp__mcp-clickhouse__list_tables {'database': 'uk'}
🤖 太好了！`uk_price_paid` 表中有超过 3000 万条房产交易记录。让我找点有意思的内容：
🛠️ mcp__mcp-clickhouse__run_select_query {'query': "\nSELECT \n    street,\n    town,\n    max(price) as max_price,\n    min(price) as min_price,\n    max(price) - min(price) as price_difference,\n    count() as sales_count\nFROM uk.uk_price_paid\nWHERE street != ''\nGROUP BY street, town\nHAVING sales_count > 100\nORDER BY price_difference DESC\nLIMIT 1\n"}
🤖 这里有个很有意思的发现：**伦敦的 Baker Street**（没错，就是福尔摩斯那条著名的街道！）在所有成交超过 100 笔的街道中，房价区间最大——最低成交价仅为 **£2,500**，最高则达到 **£5.943 亿**，惊人的价差超过 **5.94 亿英镑**！

这也说得通，因为 Baker Street 是伦敦最负盛名的地址之一，穿过 Marylebone 等富裕地区，而且在这个数据集中一共记录了 541 笔成交。

​前置条件

前置条件