使用 MCP¶
模型上下文协议 (MCP) 是一种开放协议,它规范了应用程序如何向语言模型提供工具和上下文。LangGraph 代理可以通过 langchain-mcp-adapters
库使用 MCP 服务器上定义的工具。
使用 MCP 工具¶
langchain-mcp-adapters
包使代理能够使用在一个或多个 MCP 服务器上定义的工具。
使用 MCP 服务器上定义的工具的代理
from langchain_mcp_adapters.client import MultiServerMCPClient
from langgraph.prebuilt import create_react_agent
client = MultiServerMCPClient(
{
"math": {
"command": "python",
# Replace with absolute path to your math_server.py file
"args": ["/path/to/math_server.py"],
"transport": "stdio",
},
"weather": {
# Ensure you start your weather server on port 8000
"url": "https://:8000/mcp",
"transport": "streamable_http",
}
}
)
tools = await client.get_tools()
agent = create_react_agent(
"anthropic:claude-3-7-sonnet-latest",
tools
)
math_response = await agent.ainvoke(
{"messages": [{"role": "user", "content": "what's (3 + 5) x 12?"}]}
)
weather_response = await agent.ainvoke(
{"messages": [{"role": "user", "content": "what is the weather in nyc?"}]}
)
使用 ToolNode 的 MCP 工具工作流
from langchain_mcp_adapters.client import MultiServerMCPClient
from langchain.chat_models import init_chat_model
from langgraph.graph import StateGraph, MessagesState, START
from langgraph.prebuilt import ToolNode
# Initialize the model
model = init_chat_model("anthropic:claude-3-5-sonnet-latest")
# Set up MCP client
client = MultiServerMCPClient(
{
"math": {
"command": "python",
# Make sure to update to the full absolute path to your math_server.py file
"args": ["./examples/math_server.py"],
"transport": "stdio",
},
"weather": {
# make sure you start your weather server on port 8000
"url": "https://:8000/mcp/",
"transport": "streamable_http",
}
}
)
tools = await client.get_tools()
# Bind tools to model
model_with_tools = model.bind_tools(tools)
# Create ToolNode
tool_node = ToolNode(tools)
def should_continue(state: MessagesState):
messages = state["messages"]
last_message = messages[-1]
if last_message.tool_calls:
return "tools"
return END
# Define call_model function
async def call_model(state: MessagesState):
messages = state["messages"]
response = await model_with_tools.ainvoke(messages)
return {"messages": [response]}
# Build the graph
builder = StateGraph(MessagesState)
builder.add_node("call_model", call_model)
builder.add_node("tools", tool_node)
builder.add_edge(START, "call_model")
builder.add_conditional_edges(
"call_model",
should_continue,
)
builder.add_edge("tools", "call_model")
# Compile the graph
graph = builder.compile()
# Test the graph
math_response = await graph.ainvoke(
{"messages": [{"role": "user", "content": "what's (3 + 5) x 12?"}]}
)
weather_response = await graph.ainvoke(
{"messages": [{"role": "user", "content": "what is the weather in nyc?"}]}
)
自定义 MCP 服务器¶
要创建您自己的 MCP 服务器,可以使用 mcp
库。该库提供了一种定义工具并将其作为服务器运行的简单方法。
安装 MCP 库
使用以下参考实现来测试您的代理与 MCP 工具服务器的连接。数学服务器示例 (stdio 传输)
from mcp.server.fastmcp import FastMCP
mcp = FastMCP("Math")
@mcp.tool()
def add(a: int, b: int) -> int:
"""Add two numbers"""
return a + b
@mcp.tool()
def multiply(a: int, b: int) -> int:
"""Multiply two numbers"""
return a * b
if __name__ == "__main__":
mcp.run(transport="stdio")
天气服务器示例 (可流式传输 HTTP 传输)
from mcp.server.fastmcp import FastMCP
mcp = FastMCP("Weather")
@mcp.tool()
async def get_weather(location: str) -> str:
"""Get weather for location."""
return "It's always sunny in New York"
if __name__ == "__main__":
mcp.run(transport="streamable-http")