LangChain 核心组件之Agents

原创已于 2025-11-13 01:34:56 修改 · 162 阅读

0 ·

CC 4.0 BY-SA版权

文章标签：

#langchain

于 2025-11-06 01:54:00 首次发布

Agent的定义：

Agents combine language models with tools to create systems that can reason about tasks, decide which tools to use, and iteratively work towards solutions.

Agent 可以将大语言模型跟工具结合起来，创建能够对任务进行推理、决定使用哪种工具并进行迭代的寻找解决方案的系统。

create_agent 提供 agent 实现的接口

核心组件

模型（model)

静态模型

# static model

# 方式一
from langchain.agents import create_agent
from langchain.tools import tool

@tool
def tools():
    pass
agent = create_agent(
    'gpt-5',
    tools=[tool]
)

# 方式二
from langchain.agents import create_agent
from langchain_openai import ChatOpenAI
model = ChatOpenAI(
    model='gpt-5',
    temperature=0.1,
    max_tokens=1000,
    timeout=30
)
agent1 = create_agent(model, tools=[tools])

动态模型

# dynamic model
import os
from langchain_openai import ChatOpenAI
from langchain.agents import create_agent
from langchain.agents.middleware import wrap_model_call, ModelRequest,ModelResponse

os.environ['DEEPSEEK_API_KEY'] = 'xxxxxxxxxxx'


basic_model = ChatOpenAI(
    model='deepseek-chat',
    api_key=os.getenv('DEEPSEEK_API_KEY'),
    base_url='https://api.deepseek.com',
    temperature=0.3
)

advanced_model = ChatOpenAI(
    model='deepseek-chat',
    api_key=os.getenv('DEEPSEEK_API_KEY'),
    base_url='https://api.deepseek.com',
    max_tokens=2000
)

@wrap_model_call
def dynamic_model_selection(request: ModelRequest, handler) -> ModelResponse:
    message_count = len(request.state['messages'])

    if message_count > 10:
        model = advanced_model
    else:
        model = basic_model

    request.model = model
    return handler(request)

agent = create_agent(
    model = basic_model,
    middleware=[dynamic_model_selection]
)

工具（tools）

Multiple tool calls in sequence (triggered by a single prompt)
Parallel tool calls when appropriate
Dynamic tool selection based on previous results
Tool retry logic and error handling
State persistence across tool calls

定义工具

from langchain.tools import tool
from langchain.agents import create_agent
import os
from langchain_openai import ChatOpenAI

@tool
def search(query: str) -> str:
    """Search for information."""
    return f"Results for query: {query}"

@tool
def get_weather(location: str) -> str:
    """Get weather information for a location."""
    return f"Weather in {location}: Sunny, 72F"

os.environ['DEEPSEEK_API_KEY'] = 'xxxxxxxxx'


model = ChatOpenAI(
    model='deepseek-chat',
    api_key=os.getenv('DEEPSEEK_API_KEY'),
    base_url='https://api.deepseek.com',
    temperature=0.3
)

agent = create_agent(model, tools=[search, get_weather])

工具错误处理

from langchain.tools import tool
from langchain.agents import create_agent
import os
from langchain_openai import ChatOpenAI

@tool
def search(query: str) -> str:
    """Search for information."""
    return f"Results for query: {query}"

@tool
def get_weather(location: str) -> str:
    """Get weather information for a location."""
    return f"Weather in {location}: Sunny, 72F"

os.environ['DEEPSEEK_API_KEY'] = 'xxxxxxxxx'


model = ChatOpenAI(
    model='deepseek-chat',
    api_key=os.getenv('DEEPSEEK_API_KEY'),
    base_url='https://api.deepseek.com',
    temperature=0.3
)


# Tool error handing
@wrap_tool_call
def handle_tool_errors(request, handler):
    try:
        return handler(request)
    except Exception as e:
        return ToolMessage(
            content=f"Tool Error: Please check  input and try again. ({str(e)})",
            tool_call_id=request.tool_call["id"]
        )

agent = create_agent(
    model=model,
    tools=[search, get_weather],
    middleware=[handle_tool_errors]
)

使用ReAct loop 工具

系统提示词（System prompt）

设置方式

agent = create_agent(
    model=model,
    tools=[search, get_weather],
    system_prompt="You are a helpful assistant. Be concise and accurate."
)

动态系统提示词

from typing import TypedDict
from langchain.agents import create_agent
from langchain.agents.middleware import dynamic_prompt, ModelRequest

class Context(TypedDict):
    user_role: str

@dynamic_prompt
def user_role_prompt(request: ModelRequest) -> str:
    user_role = request.runtime.context.get('user_role','user')
    base_prompt = "You are a helpful assistant."

    if user_role == 'expert':
        return f"{base_prompt} Provide detailed technical responses."
    elif user_role == 'beginner':
        return f"{base_prompt} Explain concepts simply and avoid jargon."

    return base_prompt

agent = create_agent(
    model=model,
    tools=[search],
    middleware=[user_role_prompt],
    context_schema=Context
)

result = agent.invoke(
    {"message": [{
        "role": "user",
        "content": "Explain machine learning",
    }]},
    context={"user_role": "expert"}
)

调用方式

result = agent.invoke(
    {"messages": [{"role": "user", "content": "What's the weather in San Francisco?"}]}
)