langchain-use

# LangChain Use Skill

LangChain 是构建 LLM 驱动的智能体和应用程序的开源框架。

## 安装

使用 uv 安装 LangChain（推荐，需要 Python 3.10+）：

```bash
# 安装核心包
uv add langchain

# 安装模型提供商集成
uv add langchain-anthropic  # Anthropic/Claude
uv add langchain-openai     # OpenAI
```

## 快速参考

### 核心工作流程

```
用户查询 -> create_agent() -> ReAct 循环 -> Tool 调用 -> 返回结果
```

### 创建 Agent

详见 [Agent 基础](references/agents/agent-basics.md)

```python
from langchain.agents import create_agent

agent = create_agent(
    model="claude-sonnet-4-5-20250929",
    tools=[get_weather],
    system_prompt="You are a helpful assistant",
)

# 运行 agent
result = agent.invoke(
    {"messages": [{"role": "user", "content": "what is the weather in sf"}]}
)
```

### 定义 Tool

详见 [Tool 基础](references/tools/tool-basics.md)

```python
from langchain.tools import tool

@tool
def get_weather(city: str) -> str:
    """Get weather for a given city."""
    return f"It's always sunny in {city}!"
```

### 访问 Runtime Context

使用 `ToolRuntime` 访问 state、context、store：

```python
from langchain.tools import tool, ToolRuntime
from dataclasses import dataclass

@dataclass
class Context:
    user_id: str

@tool
def get_user_location(runtime: ToolRuntime[Context]) -> str:
    """Retrieve user location based on user ID."""
    user_id = runtime.context.user_id
    return "Florida" if user_id == "1" else "SF"
```

### 管理 Memory

详见 [短期记忆](references/memory/short-term-memory.md)
详见 [长期记忆](references/memory/long-term-memory.md)

```python
from langgraph.checkpoint.memory import InMemorySaver

agent = create_agent(
    model,
    tools,
    checkpointer=InMemorySaver(),  # 短期记忆
)

# 使用 thread_id 维护会话
config = {"configurable": {"thread_id": "1"}}
agent.invoke({"messages": [...]}, config)
```

### 添加 Middleware

详见 [中间件概述](references/middleware/middleware-overview.md)

```python
from langchain.agents.middleware import before_model, after_model

@before_model
def trim_messages(state, runtime):
    # 消息修剪逻辑
    return None
```

## 进阶主题

| 主题 | 文档 | 说明 |
|------|--