构建自定义 SQL 代理

在本教程中，我们将构建一个自定义代理，该代理能够使用 LangGraph 回答关于 SQL 数据库的问题。 LangChain 提供了内置的代理实现，这些实现使用 LangGraph 原语构建。如果需要更深层次的定制，可以直接在 LangGraph 中实现代理。本指南演示了一个 SQL 代理的实现示例。有关实践入门，请参阅使用更高级的 LangChain 抽象构建 SQL 代理。

构建 SQL 数据库的问答系统需要执行模型生成的 SQL 查询。这样做存在固有风险。请确保您的数据库连接权限始终尽可能严格地限定在代理所需范围内。这将减轻（但不能消除）构建模型驱动系统的风险。

预构建代理让我们可以快速上手，但我们依赖系统提示来约束其行为——例如，我们指示代理始终从“列出表”工具开始，并在执行查询前始终运行查询检查工具。我们可以通过自定义代理在 LangGraph 中实施更高程度的控制。在这里，我们实现了一个简单的 ReAct 代理设置，为特定的工具调用设置了专用节点。我们将使用与预构建代理相同的[状态]。

概念

我们将涵盖以下概念：

用于从 SQL 数据库读取的工具
LangGraph 图 API，包括状态、节点、边和条件边。
Human in the Loop流程

设置

安装

pip install langchain  langgraph  langchain-community

LangSmith

设置 LangSmith 以检查链或代理内部发生的情况。然后设置以下环境变量：

export LANGSMITH_TRACING="true"
export LANGSMITH_API_KEY="..."

1. 选择一个 LLM

选择一个支持工具调用的模型：

👉 阅读 OpenAI 聊天模型集成文档

pip install -U "langchain[openai]"

import os
from langchain.chat_models import init_chat_model

os.environ["OPENAI_API_KEY"] = "sk-..."

model = init_chat_model("gpt-5.4")

import os
from langchain_openai import ChatOpenAI

os.environ["OPENAI_API_KEY"] = "sk-..."

model = ChatOpenAI(model="gpt-5.4")

👉 阅读 Anthropic 聊天模型集成文档

pip install -U "langchain[anthropic]"

import os
from langchain.chat_models import init_chat_model

os.environ["ANTHROPIC_API_KEY"] = "sk-..."

model = init_chat_model("claude-sonnet-4-6")

import os
from langchain_anthropic import ChatAnthropic

os.environ["ANTHROPIC_API_KEY"] = "sk-..."

model = ChatAnthropic(model="claude-sonnet-4-6")

👉 阅读 Azure 聊天模型集成文档

pip install -U "langchain[openai]"

import os
from langchain.chat_models import init_chat_model

os.environ["AZURE_OPENAI_API_KEY"] = "..."
os.environ["AZURE_OPENAI_ENDPOINT"] = "..."
os.environ["OPENAI_API_VERSION"] = "2025-03-01-preview"

model = init_chat_model(
    "azure_openai:gpt-5.4",
    azure_deployment=os.environ["AZURE_OPENAI_DEPLOYMENT_NAME"],
)

import os
from langchain_openai import AzureChatOpenAI

os.environ["AZURE_OPENAI_API_KEY"] = "..."
os.environ["AZURE_OPENAI_ENDPOINT"] = "..."
os.environ["OPENAI_API_VERSION"] = "2025-03-01-preview"

model = AzureChatOpenAI(
    model="gpt-5.4",
    azure_deployment=os.environ["AZURE_OPENAI_DEPLOYMENT_NAME"]
)

👉 阅读 Google GenAI 聊天模型集成文档

pip install -U "langchain[google-genai]"

import os
from langchain.chat_models import init_chat_model

os.environ["GOOGLE_API_KEY"] = "..."

model = init_chat_model("google_genai:gemini-2.5-flash-lite")

import os
from langchain_google_genai import ChatGoogleGenerativeAI

os.environ["GOOGLE_API_KEY"] = "..."

model = ChatGoogleGenerativeAI(model="gemini-2.5-flash-lite")

👉 阅读 AWS Bedrock 聊天模型集成文档

pip install -U "langchain[aws]"

from langchain.chat_models import init_chat_model

# 按照此处步骤配置您的凭证：
# https://docs.aws.amazon.com/bedrock/latest/userguide/getting-started.html

model = init_chat_model(
    "anthropic.claude-3-5-sonnet-20240620-v1:0",
    model_provider="bedrock_converse",
)

from langchain_aws import ChatBedrock

model = ChatBedrock(model="anthropic.claude-3-5-sonnet-20240620-v1:0")

👉 阅读 HuggingFace 聊天模型集成文档

pip install -U "langchain[huggingface]"

import os
from langchain.chat_models import init_chat_model

os.environ["HUGGINGFACEHUB_API_TOKEN"] = "hf_..."

model = init_chat_model(
    "microsoft/Phi-3-mini-4k-instruct",
    model_provider="huggingface",
    temperature=0.7,
    max_tokens=1024,
)

import os
from langchain_huggingface import ChatHuggingFace, HuggingFaceEndpoint

os.environ["HUGGINGFACEHUB_API_TOKEN"] = "hf_..."

llm = HuggingFaceEndpoint(
    repo_id="microsoft/Phi-3-mini-4k-instruct",
    temperature=0.7,
    max_length=1024,
)
model = ChatHuggingFace(llm=llm)

👉 阅读 OpenRouter 聊天模型集成文档

pip install -U "langchain-openrouter"

import os
from langchain.chat_models import init_chat_model

os.environ["OPENROUTER_API_KEY"] = "sk-..."

model = init_chat_model(
    "auto",
    model_provider="openrouter",
)

import os
from langchain_openrouter import ChatOpenRouter

os.environ["OPENROUTER_API_KEY"] = "sk-..."

model = ChatOpenRouter(model="auto")

下面示例中显示的输出使用了 OpenAI。

2. 配置数据库

您将为本教程创建一个 SQLite 数据库。SQLite 是一个轻量级数据库，易于设置和使用。我们将加载 chinook 数据库，这是一个代表数字媒体商店的示例数据库。为方便起见，我们将数据库（Chinook.db）托管在一个公共的 GCS 存储桶中。

import requests, pathlib

url = "https://storage.googleapis.com/benchmarks-artifacts/chinook/Chinook.db"
local_path = pathlib.Path("Chinook.db")

if local_path.exists():
    print(f"{local_path} already exists, skipping download.")
else:
    response = requests.get(url)
    if response.status_code == 200:
        local_path.write_bytes(response.content)
        print(f"File downloaded and saved as {local_path}")
    else:
        print(f"Failed to download the file. Status code: {response.status_code}")

我们将使用 langchain_community 包中提供的便捷 SQL 数据库包装器与数据库交互。该包装器提供了一个简单的接口来执行 SQL 查询并获取结果：

from langchain_community.utilities import SQLDatabase

db = SQLDatabase.from_uri("sqlite:///Chinook.db")

print(f"Dialect: {db.dialect}")
print(f"Available tables: {db.get_usable_table_names()}")
print(f'Sample output: {db.run("SELECT * FROM Artist LIMIT 5;")}')

Dialect: sqlite
Available tables: ['Album', 'Artist', 'Customer', 'Employee', 'Genre', 'Invoice', 'InvoiceLine', 'MediaType', 'Playlist', 'PlaylistTrack', 'Track']
Sample output: [(1, 'AC/DC'), (2, 'Accept'), (3, 'Aerosmith'), (4, 'Alanis Morissette'), (5, 'Alice In Chains')]

3. 添加用于数据库交互的工具

使用 langchain_community 包中提供的 SQLDatabase 包装器与数据库交互。该包装器提供了一个简单的接口来执行 SQL 查询并获取结果：

from langchain_community.agent_toolkits import SQLDatabaseToolkit

toolkit = SQLDatabaseToolkit(db=db, llm=model)

tools = toolkit.get_tools()

for tool in tools:
    print(f"{tool.name}: {tool.description}\n")

sql_db_query: Input to this tool is a detailed and correct SQL query, output is a result from the database. If the query is not correct, an error message will be returned. If an error is returned, rewrite the query, check the query, and try again. If you encounter an issue with Unknown column 'xxxx' in 'field list', use sql_db_schema to query the correct table fields.

sql_db_schema: Input to this tool is a comma-separated list of tables, output is the schema and sample rows for those tables. Be sure that the tables actually exist by calling sql_db_list_tables first! Example Input: table1, table2, table3

sql_db_list_tables: Input is an empty string, output is a comma-separated list of tables in the database.

sql_db_query_checker: Use this tool to double check if your query is correct before executing it. Always use this tool before executing a query with sql_db_query!

4. 定义应用步骤

我们为以下步骤构建专用节点：

列出数据库表
调用“获取模式”工具
生成查询
检查查询

将这些步骤放在专用节点中，使我们能够 (1) 在需要时强制进行工具调用，以及 (2) 自定义与每个步骤关联的提示。

from typing import Literal

from langchain.messages import AIMessage
from langchain_core.runnables import RunnableConfig
from langgraph.graph import END, START, MessagesState, StateGraph
from langgraph.prebuilt import ToolNode


get_schema_tool = next(tool for tool in tools if tool.name == "sql_db_schema")
get_schema_node = ToolNode([get_schema_tool], name="get_schema")

run_query_tool = next(tool for tool in tools if tool.name == "sql_db_query")
run_query_node = ToolNode([run_query_tool], name="run_query")


# 示例：创建一个预定的工具调用
def list_tables(state: MessagesState):
    tool_call = {
        "name": "sql_db_list_tables",
        "args": {},
        "id": "abc123",
        "type": "tool_call",
    }
    tool_call_message = AIMessage(content="", tool_calls=[tool_call])

    list_tables_tool = next(tool for tool in tools if tool.name == "sql_db_list_tables")
    tool_message = list_tables_tool.invoke(tool_call)
    response = AIMessage(f"Available tables: {tool_message.content}")

    return {"messages": [tool_call_message, tool_message, response]}


# 示例：强制模型创建工具调用
def call_get_schema(state: MessagesState):
    # 注意，LangChain 强制所有模型接受 `tool_choice="any"`
    # 以及 `tool_choice=<工具名称字符串>`。
    llm_with_tools = model.bind_tools([get_schema_tool], tool_choice="any")
    response = llm_with_tools.invoke(state["messages"])

    return {"messages": [response]}


generate_query_system_prompt = """
You are an agent designed to interact with a SQL database.
Given an input question, create a syntactically correct {dialect} query to run,
then look at the results of the query and return the answer. Unless the user
specifies a specific number of examples they wish to obtain, always limit your
query to at most {top_k} results.

You can order the results by a relevant column to return the most interesting
examples in the database. Never query for all the columns from a specific table,
only ask for the relevant columns given the question.

DO NOT make any DML statements (INSERT, UPDATE, DELETE, DROP etc.) to the database.
""".format(
    dialect=db.dialect,
    top_k=5,
)


def generate_query(state: MessagesState):
    system_message = {
        "role": "system",
        "content": generate_query_system_prompt,
    }
    # 我们在这里不强制工具调用，以允许模型在获得解决方案时自然响应。
    llm_with_tools = model.bind_tools([run_query_tool])
    response = llm_with_tools.invoke([system_message] + state["messages"])

    return {"messages": [response]}


check_query_system_prompt = """
You are a SQL expert with a strong attention to detail.
Double check the {dialect} query for common mistakes, including:
- Using NOT IN with NULL values
- Using UNION when UNION ALL should have been used
- Using BETWEEN for exclusive ranges
- Data type mismatch in predicates
- Properly quoting identifiers
- Using the correct number of arguments for functions
- Casting to the correct data type
- Using the proper columns for joins

If there are any of the above mistakes, rewrite the query. If there are no mistakes,
just reproduce the original query.

You will call the appropriate tool to execute the query after running this check.
""".format(dialect=db.dialect)


def check_query(state: MessagesState):
    system_message = {
        "role": "system",
        "content": check_query_system_prompt,
    }

    # 生成一个用于检查的人工用户消息
    tool_call = state["messages"][-1].tool_calls[0]
    user_message = {"role": "user", "content": tool_call["args"]["query"]}
    llm_with_tools = model.bind_tools([run_query_tool], tool_choice="any")
    response = llm_with_tools.invoke([system_message, user_message])
    response.id = state["messages"][-1].id

    return {"messages": [response]}

5. 实现代理

我们现在可以使用图 API 将这些步骤组装成一个工作流。我们在查询生成步骤定义了一个条件边，如果生成了查询，则路由到查询检查器；如果没有工具调用（即 LLM 已对查询做出响应），则结束。

def should_continue(state: MessagesState) -> Literal[END, "check_query"]:
    messages = state["messages"]
    last_message = messages[-1]
    if not last_message.tool_calls:
        return END
    else:
        return "check_query"


builder = StateGraph(MessagesState)
builder.add_node(list_tables)
builder.add_node(call_get_schema)
builder.add_node(get_schema_node, "get_schema")
builder.add_node(generate_query)
builder.add_node(check_query)
builder.add_node(run_query_node, "run_query")

builder.add_edge(START, "list_tables")
builder.add_edge("list_tables", "call_get_schema")
builder.add_edge("call_get_schema", "get_schema")
builder.add_edge("get_schema", "generate_query")
builder.add_conditional_edges(
    "generate_query",
    should_continue,
)
builder.add_edge("check_query", "run_query")
builder.add_edge("run_query", "generate_query")

agent = builder.compile()

我们在下面可视化该应用程序：

from IPython.display import Image, display
from langchain_core.runnables.graph import CurveStyle, MermaidDrawMethod, NodeStyles

display(Image(agent.get_graph().draw_mermaid_png()))

我们现在可以调用该图：

question = "Which genre on average has the longest tracks?"

for step in agent.stream(
    {"messages": [{"role": "user", "content": question}]},
    stream_mode="values",
):
    step["messages"][-1].pretty_print()

================================ Human Message =================================

Which genre on average has the longest tracks?
================================== Ai Message ==================================

Available tables: Album, Artist, Customer, Employee, Genre, Invoice, InvoiceLine, MediaType, Playlist, PlaylistTrack, Track
================================== Ai Message ==================================
Tool Calls:
  sql_db_schema (call_yzje0tj7JK3TEzDx4QnRR3lL)
 Call ID: call_yzje0tj7JK3TEzDx4QnRR3lL
  Args:
    table_names: Genre, Track
================================= Tool Message =================================
Name: sql_db_schema


CREATE TABLE "Genre" (
	"GenreId" INTEGER NOT NULL,
	"Name" NVARCHAR(120),
	PRIMARY KEY ("GenreId")
)

/*
3 rows from Genre table:
GenreId	Name
1	Rock
2	Jazz
3	Metal
*/


CREATE TABLE "Track" (
	"TrackId" INTEGER NOT NULL,
	"Name" NVARCHAR(200) NOT NULL,
	"AlbumId" INTEGER,
	"MediaTypeId" INTEGER NOT NULL,
	"GenreId" INTEGER,
	"Composer" NVARCHAR(220),
	"Milliseconds" INTEGER NOT NULL,
	"Bytes" INTEGER,
	"UnitPrice" NUMERIC(10, 2) NOT NULL,
	PRIMARY KEY ("TrackId"),
	FOREIGN KEY("MediaTypeId") REFERENCES "MediaType" ("MediaTypeId"),
	FOREIGN KEY("GenreId") REFERENCES "Genre" ("GenreId"),
	FOREIGN KEY("AlbumId") REFERENCES "Album" ("AlbumId")
)

/*
3 rows from Track table:
TrackId	Name	AlbumId	MediaTypeId	GenreId	Composer	Milliseconds	Bytes	UnitPrice
1	For Those About To Rock (We Salute You)	1	1	1	Angus Young, Malcolm Young, Brian Johnson	343719	11170334	0.99
2	Balls to the Wall	2	2	1	U. Dirkschneider, W. Hoffmann, H. Frank, P. Baltes, S. Kaufmann, G. Hoffmann	342562	5510424	0.99
3	Fast As a Shark	3	2	1	F. Baltes, S. Kaufman, U. Dirkscneider & W. Hoffman	230619	3990994	0.99
*/
================================== Ai Message ==================================
Tool Calls:
  sql_db_query (call_cb9ApLfZLSq7CWg6jd0im90b)
 Call ID: call_cb9ApLfZLSq7CWg6jd0im90b
  Args:
    query: SELECT Genre.Name, AVG(Track.Milliseconds) AS AvgMilliseconds FROM Track JOIN Genre ON Track.GenreId = Genre.GenreId GROUP BY Genre.GenreId ORDER BY AvgMilliseconds DESC LIMIT 5;
================================== Ai Message ==================================
Tool Calls:
  sql_db_query (call_DMVALfnQ4kJsuF3Yl6jxbeAU)
 Call ID: call_DMVALfnQ4kJsuF3Yl6jxbeAU
  Args:
    query: SELECT Genre.Name, AVG(Track.Milliseconds) AS AvgMilliseconds FROM Track JOIN Genre ON Track.GenreId = Genre.GenreId GROUP BY Genre.GenreId ORDER BY AvgMilliseconds DESC LIMIT 5;
================================= Tool Message =================================
Name: sql_db_query

[('Sci Fi & Fantasy', 2911783.0384615385), ('Science Fiction', 2625549.076923077), ('Drama', 2575283.78125), ('TV Shows', 2145041.0215053763), ('Comedy', 1585263.705882353)]
================================== Ai Message ==================================

The genre with the longest tracks on average is "Sci Fi & Fantasy," with an average track length of approximately 2,911,783 milliseconds. Other genres with relatively long tracks include "Science Fiction," "Drama," "TV Shows," and "Comedy."

有关上述运行，请参阅 LangSmith 跟踪。

6. 实现Human in the Loop审查

在执行代理的 SQL 查询之前进行检查是谨慎的做法，以避免任何意外操作或低效。这里我们利用 LangGraph 的Human in the Loop功能，在执行 SQL 查询前暂停运行并等待人工审查。使用 LangGraph 的持久化层，我们可以无限期地暂停运行（或至少在持久化层存活期间）。让我们将 sql_db_query 工具包装在一个接收人工输入的节点中。我们可以使用 interrupt 函数来实现这一点。下面，我们允许输入以批准工具调用、编辑其参数或提供用户反馈。

from langchain_core.runnables import RunnableConfig
from langchain.tools import tool
from langgraph.types import interrupt

@tool(
    run_query_tool.name,
    description=run_query_tool.description,
    args_schema=run_query_tool.args_schema
)
def run_query_tool_with_interrupt(config: RunnableConfig, **tool_input):
    request = {
        "action": run_query_tool.name,
        "args": tool_input,
        "description": "Please review the tool call"
    }
    response = interrupt([request])
    # 批准工具调用
    if response["type"] == "accept":
        tool_response = run_query_tool.invoke(tool_input, config)
    # 更新工具调用参数
    elif response["type"] == "edit":
        tool_input = response["args"]["args"]
        tool_response = run_query_tool.invoke(tool_input, config)
    # 用用户反馈响应 LLM
    elif response["type"] == "response":
        user_feedback = response["args"]
        tool_response = user_feedback
    else:
        raise ValueError(f"Unsupported interrupt response type: {response['type']}")

    return tool_response

# 重新定义工具节点以使用中断版本
run_query_node = ToolNode([run_query_tool_with_interrupt], name="run_query")

上述实现遵循了更广泛的Human in the Loop指南中的工具中断示例。有关详细信息和替代方案，请参阅该指南。

现在让我们重新组装我们的图。我们将用人工审查替换程序化检查。请注意，我们现在包含了一个检查点；这是暂停和恢复运行所必需的。

from langgraph.checkpoint.memory import InMemorySaver

def should_continue(state: MessagesState) -> Literal[END, "run_query"]:
    messages = state["messages"]
    last_message = messages[-1]
    if not last_message.tool_calls:
        return END
    else:
        return "run_query"

builder = StateGraph(MessagesState)
builder.add_node(list_tables)
builder.add_node(call_get_schema)
builder.add_node(get_schema_node, "get_schema")
builder.add_node(generate_query)
builder.add_node(run_query_node, "run_query")

builder.add_edge(START, "list_tables")
builder.add_edge("list_tables", "call_get_schema")
builder.add_edge("call_get_schema", "get_schema")
builder.add_edge("get_schema", "generate_query")
builder.add_conditional_edges(
    "generate_query",
    should_continue,
)
builder.add_edge("run_query", "generate_query")

checkpointer = InMemorySaver()
agent = builder.compile(checkpointer=checkpointer)

我们可以像以前一样调用该图。这次，执行被中断了：

import json

config = {"configurable": {"thread_id": "1"}}

question = "Which genre on average has the longest tracks?"

for step in agent.stream(
    {"messages": [{"role": "user", "content": question}]},
    config,
    stream_mode="values",
):
    if "messages" in step:
        step["messages"][-1].pretty_print()
    elif "__interrupt__" in step:
        action = step["__interrupt__"][0]
        print("INTERRUPTED:")
        for request in action.value:
            print(json.dumps(request, indent=2))
    else:
        pass

...

INTERRUPTED:
{
  "action": "sql_db_query",
  "args": {
    "query": "SELECT Genre.Name, AVG(Track.Milliseconds) AS AvgLength FROM Track JOIN Genre ON Track.GenreId = Genre.GenreId GROUP BY Genre.Name ORDER BY AvgLength DESC LIMIT 5;"
  },
  "description": "Please review the tool call"
}

我们可以使用 Command 来接受或编辑工具调用：

from langgraph.types import Command


for step in agent.stream(
    Command(resume={"type": "accept"}),
    # Command(resume={"type": "edit", "args": {"query": "..."}}),
    config,
    stream_mode="values",
):
    if "messages" in step:
        step["messages"][-1].pretty_print()
    elif "__interrupt__" in step:
        action = step["__interrupt__"][0]
        print("INTERRUPTED:")
        for request in action.value:
            print(json.dumps(request, indent=2))
    else:
        pass

================================== Ai Message ==================================
Tool Calls:
  sql_db_query (call_t4yXkD6shwdTPuelXEmY3sAY)
 Call ID: call_t4yXkD6shwdTPuelXEmY3sAY
  Args:
    query: SELECT Genre.Name, AVG(Track.Milliseconds) AS AvgLength FROM Track JOIN Genre ON Track.GenreId = Genre.GenreId GROUP BY Genre.Name ORDER BY AvgLength DESC LIMIT 5;
================================= Tool Message =================================
Name: sql_db_query

[('Sci Fi & Fantasy', 2911783.0384615385), ('Science Fiction', 2625549.076923077), ('Drama', 2575283.78125), ('TV Shows', 2145041.0215053763), ('Comedy', 1585263.705882353)]
================================== Ai Message ==================================

The genre with the longest average track length is "Sci Fi & Fantasy" with an average length of about 2,911,783 milliseconds. Other genres with long average track lengths include "Science Fiction," "Drama," "TV Shows," and "Comedy."

有关详细信息，请参阅Human in the Loop指南。

后续步骤

查看评估图指南，了解如何使用 LangSmith 评估 LangGraph 应用程序，包括像这样的 SQL 代理。

将这些文档通过 MCP 连接到 Claude、VSCode 等，以获取实时答案。

在 GitHub 上编辑此页面或提交问题。

Tutorials

Conceptual overviews

Additional resources

概念

设置

安装

LangSmith

1. 选择一个 LLM

2. 配置数据库

3. 添加用于数据库交互的工具

4. 定义应用步骤

5. 实现代理

6. 实现Human in the Loop审查

后续步骤

​概念

​设置

​安装

​LangSmith

​1. 选择一个 LLM

​2. 配置数据库

​3. 添加用于数据库交互的工具

​4. 定义应用步骤

​5. 实现代理

​6. 实现Human in the Loop审查

​后续步骤

概念

设置

安装

LangSmith

1. 选择一个 LLM

2. 配置数据库

3. 添加用于数据库交互的工具

4. 定义应用步骤

5. 实现代理

6. 实现Human in the Loop审查

后续步骤