MCP
基本介绍
官方地址：
- https://modelcontextprotocol.io/introduction
“MCP 是一种开放协议，旨在标准化应用程序向大型语言模型（LLM）提供上下文的方式。可以把 MCP 想象成 AI 应用程序的 USB-C 接口。就像 USB-C 提供了一种标准化的方式，让你的设备能够连接各种外设和配件一样，MCP 也提供了一种标准化的方式，让 AI 模型能够连接不同的数据源和工具。”
● MCP 主机（MCP Hosts）：像 Claude Desktop、IDE 或 AI 工具等程序，它们希望通过 MCP 访问数据。
● MCP 客户端（MCP Clients）：维护与服务器 1:1 连接的协议客户端。
● MCP 服务器（MCP Servers）：轻量级程序，它们通过标准化的模型上下文协议（Model Context Protocol）公开特定的功能。
● 本地数据源（Local Data Sources）：你的计算机上的文件、数据库和服务，MCP 服务器可以安全地访问这些数据。
● 远程服务（Remote Services）：通过互联网可用的外部系统（例如 API），MCP 服务器可以与其连接。
https://www.anthropic.com/news/model-context-protocol
测试项目
服务端
使用刚才安装的 uv 创建一个新的项目，并启动虚拟环境
uv init weather
cd weather
uv venv
source .venv/bin/activate
操作结果如下：
安装依赖
uv add "mcp[cli]" httpx
对应如下：
编写一个 server：
from typing import Any
import httpx
from mcp.server.fastmcp import FastMCP
mcp = FastMCP("weather")
NWS_API_BASE = "https://api.weather.gov"
USER_AGENT = "weather-app/1.0"
# 这里是用来发起请求的
async def make_nws_request(url: str) -> dict[str, Any] | None:
"""Make a request to the NWS API with proper error handling."""
headers = {
"User-Agent": USER_AGENT,
"Accept": "application/geo+json"
}
async with httpx.AsyncClient() as client:
try:
response = await client.get(url, headers=headers, timeout=30.0)
response.raise_for_status()
return response.json()
except Exception:
return None
# 模板格式 将结果对应的内容 补充进模板
def format_alert(feature: dict) -> str:
"""Format an alert feature into a readable string."""
props = feature["properties"]
return f"""
Event: {props.get('event', 'Unknown')}
Area: {props.get('areaDesc', 'Unknown')}
Severity: {props.get('severity', 'Unknown')}
Description: {props.get('description', 'No description available')}
Instructions: {props.get('instruction', 'No specific instructions provided')}
"""
# 通过 mcp.tool 标识功能
# 根据 区域名称 进行查询
# 这里主要是拼接URL，并且对URL发起了请求make_nws_request
# 最后将返回结果放进format_alert格式化函数
@mcp.tool()
async def get_alerts(state: str) -> str:
"""Get weather alerts for a US state.
Args:
state: Two-letter US state code (e.g. CA, NY)
"""
url = f"{NWS_API_BASE}/alerts/active/area/{state}"
data = await make_nws_request(url)
if not data or "features" not in data:
return "Unable to fetch alerts or no alerts found."
if not data["features"]:
return "No active alerts for this state."
alerts = [format_alert(feature) for feature in data["features"]]
return "\n---\n".join(alerts)
# 通过 mcp.tool 标识功能
# 根据 lng lat 进行查询
# 主要功能类似于 get_alerts
@mcp.tool()
async def get_forecast(latitude: float, longitude: float) -> str:
"""Get weather forecast for a location.
Args:
latitude: Latitude of the location
longitude: Longitude of the location
"""
# First get the forecast grid endpoint
points_url = f"{NWS_API_BASE}/points/{latitude},{longitude}"
points_data = await make_nws_request(points_url)
if not points_data:
return "Unable to fetch forecast data for this location."
# Get the forecast URL from the points response
forecast_url = points_data["properties"]["forecast"]
forecast_data = await make_nws_request(forecast_url)
if not forecast_data:
return "Unable to fetch detailed forecast."
# Format the periods into a readable forecast
periods = forecast_data["properties"]["periods"]
forecasts = []
for period in periods[:5]: # Only show next 5 periods
forecast = f"""
{period['name']}:
Temperature: {period['temperature']}°{period['temperatureUnit']}
Wind: {period['windSpeed']} {period['windDirection']}
Forecast: {period['detailedForecast']}
"""
forecasts.append(forecast)
return "\n---\n".join(forecasts)
# 启动函数
if __name__ == "__main__":
# Initialize and run the server
mcp.run(transport='stdio')
客户端
继续新建一个项目：
uv init mcp-client
cd mcp-client
uv venv
source .venv/bin/activate
uv add mcp openai python-dotenv
对应结果如下：
编写一个 client 代码：
import asyncio
from typing import Optional
from contextlib import AsyncExitStack
from mcp import ClientSession, StdioServerParameters
from mcp.client.stdio import stdio_client
from openai import OpenAI
from dotenv import load_dotenv
import json
load_dotenv()
class MCPClient:
def __init__(self):
# 初始化的参数配置在这里
self.session: Optional[ClientSession] = None
self.exit_stack = AsyncExitStack()
self.openai = OpenAI()
self.model = "gpt-4o-mini"
# 连接到服务器
async def connect_to_server(self, server_script_path: str):
"""Connect to an MCP server
Args:
server_script_path: Path to the server script (.py or .js)
"""
# 为了兼容 py 和 js
is_python = server_script_path.endswith('.py')
is_js = server_script_path.endswith('.js')
if not (is_python or is_js):
raise ValueError("Server script must be a .py or .js file")
command = "python" if is_python else "node"
server_params = StdioServerParameters(
command=command,
args=[server_script_path],
env=None
)
stdio_transport = await self.exit_stack.enter_async_context(stdio_client(server_params))
self.stdio, self.write = stdio_transport
self.session = await self.exit_stack.enter_async_context(ClientSession(self.stdio, self.write))
await self.session.initialize()
# 从 Server 获取所有用的工具
# List available tools
response = await self.session.list_tools()
tools = response.tools
print("\nConnected to server with tools:", [tool.name for tool in tools])
# 执行查询
async def process_query(self, query: str) -> str:
"""Process a query using Claude and available tools"""
messages = [
{
"role": "user",
"content": query
}
]
# 1.从 Server 获取到所有可用的客户端后
response = await self.session.list_tools()
# 2.转换为 GPT 需要的格式
available_tools = [{
"type": "function",
"function": {
"name": tool.name,
"description": tool.description,
"parameters": tool.inputSchema
}
} for tool in response.tools]
# 3.在调用的时候 带着 tools
response = self.openai.chat.completions.create(
model=self.model,
max_tokens=1000,
messages=messages,
tools=available_tools
)
tool_results = []
final_text = []
for choice in response.choices:
message = choice.message
is_function_call = message.tool_calls
if not is_function_call:
final_text.append(message.content)
else:
tool_name = message.tool_calls[0].function.name
tool_args = json.loads(message.tool_calls[0].function.arguments)
print(f"Tool Call: {tool_name}")
print(f"Tool Call Param: {json.dumps(tool_args, ensure_ascii=False, indent=2)}")
result = await self.session.call_tool(tool_name, tool_args)
tool_results.append({"call": tool_name, "result": result})
final_text.append(f"[Calling tool {tool_name} with args {tool_args}]")
if message.content and hasattr(message.content, 'text'):
messages.append({
"role": "assistant",
"content": message.content
})
messages.append({
"role": "user",
"content": result.content
})
response = self.openai.chat.completions.create(
model=self.model,
max_tokens=1000,
messages=messages,
tools=available_tools
)
final_text.append(response.choices[0].message.content)
return "\\n".join(final_text)
async def chat_loop(self):
"""Run an interactive chat loop"""
print("\nMCP Client Started!")
print("Type your queries or 'quit' to exit.")
while True:
try:
query = input("\nQuery: ").strip()
if query.lower() == 'quit':
break
response = await self.process_query(query)
print("\n" + response)
except Exception as e:
print(f"\nError: {str(e)}")
async def cleanup(self):
"""Clean up resources"""
await self.exit_stack.aclose()
async def main():
if len(sys.argv) < 2:
print("Usage: python client.py <path_to_server_script>")
sys.exit(1)
client = MCPClient()
try:
await client.connect_to_server(sys.argv[1])
await client.chat_loop()
finally:
await client.cleanup()
if __name__ == "__main__":
import sys
asyncio.run(main())
测试项目
这里我们启动 Client 端、Server 端。
第一个参数是 Client 端，第二个是 Server 端：
uv run main.py ../weather/main.py
启动之后，可以看到从 Server 端获取到了目前有的工具：
我们提问：“What are the weather alerts in California”。
PS：注意，这里需要是美国地址，我们 Server 端中的URL是美国的。
在下面的日志中，我们可以看到 Client 中输出的：
Tool Call: get_alerts
Tool Call Param: {
"state": "CA"
}
全部的内容如下所示：
➜ uv run main.py ../weather/main.py
Connected to server with tools: ['get_alerts', 'get_forecast']
MCP Client Started!
Type your queries or 'quit' to exit.
Query: What are the weather alerts in California
Tool Call: get_alerts
Tool Call Param: {
"state": "CA"
}
[Calling tool get_alerts with args {'state': 'CA'}]\nHere are the current weather alerts in California:
1. **Lake Wind Advisory**
- **Area**: Greater Lake Tahoe Area
- **Severity**: Moderate
- **Description**: Southwest winds 15 to 25 mph with gusts up to 45 mph and waves 1 to 4 feet.
- **When**: Until 5 AM PDT Thursday.
- **Impacts**: Small boats, kayaks, and paddle boards will be prone to capsizing and should stay off lake waters until conditions improve.
- **Instructions**: Check lake conditions before heading out and consider postponing boating activities until a day with less wind.
2. **Frost Advisory**
- **Area**: Southeastern Mendocino Interior; Southern Lake County
- **Severity**: Minor
- **Description**: Temperatures as low as 33 will result in frost formation.
- **When**: From 2 AM to 10 AM PDT Thursday.
- **Impacts**: Frost could harm sensitive outdoor vegetation.
- **Instructions**: Take steps now to protect tender plants from the cold.
3. **Freeze Warning**
- **Area**: Northeastern Mendocino Interior
- **Severity**: Moderate
- **Description**: Sub-freezing temperatures as low as 30 expected.
- **When**: From 3 AM to 10 AM PDT Thursday.
- **Impacts**: Frost/Freeze damage to crops.
- **Instructions**: Take steps now to protect tender plants from the cold.
4. **Freeze Warning**
- **Area**: Northwestern Mendocino Interior
- **Severity**: Moderate
- **Description**: Sub-freezing temperatures as low as 29 expected.
- **When**: From 3 AM to 10 AM PDT Thursday.
- **Impacts**: Frost/Freeze damage to crops.
- **Instructions**: Take steps now to protect tender plants from the cold.
5. **Freeze Warning**
- **Area**: Northern Lake County
- **Severity**: Moderate
- **Description**: Sub-freezing temperatures as low as 29 expected.
- **When**: From 3 AM to 10 AM PDT Thursday.
- **Impacts**: Frost/Freeze damage to crops.
- **Instructions**: Take steps now to protect tender plants from the cold.
6. **Winter Weather Advisory**
- **Area**: Northern Trinity
- **Severity**: Moderate
- **Description**: Snow above 4000 feet, with additional snow accumulations up to one inch.
- **When**: Until 1 AM PDT Thursday.
- **Impacts**: Plan on slippery road conditions.
- **Instructions**: Slow down and use caution while traveling.
7. **Winter Weather Advisory**
- **Area**: Western Siskiyou County
- **Severity**: Moderate
- **Description**: Snow expected, with total accumulations between 6 and 8 inches and up to 12 inches over high remote terrain. Winds gusting as high as 55 mph.
- **When**: Until 11 AM PDT Thursday.
- **Impacts**: Travel could be difficult, affecting commutes.
- **Instructions**: Slow down and use caution while traveling. Call 511 or visit quickmap.dot.ca.gov for road information.
Please take necessary precautions based on these alerts!
Query:
对应的截图如下所示：
