121 lines
3.8 KiB
Python
121 lines
3.8 KiB
Python
import json
|
|
import os
|
|
import asyncio
|
|
|
|
# --- 核心：导入官方库 ---
|
|
from langchain_mcp_adapters.client import MultiServerMCPClient
|
|
|
|
# LangChain/LangGraph 组件
|
|
from langchain_openai import ChatOpenAI
|
|
from langchain_core.messages import SystemMessage, ToolMessage
|
|
from langgraph.graph import StateGraph, MessagesState, START, END
|
|
from langgraph.prebuilt import ToolNode
|
|
|
|
# 复用你的流式输出模块和配置
|
|
from m11_mcp_advanced.s01_agent_stream import run_agent_with_streaming
|
|
from config import OPENAI_API_KEY, AMAP_MAPS_API_KEY
|
|
|
|
# === 配置 MCP 服务器 ===
|
|
MCP_SERVERS = {
|
|
# 方式1.1: 云端代理 —— stdio模式
|
|
"高德地图": {
|
|
"transport": "stdio",
|
|
"command": "npx",
|
|
"args": ["-y", "@amap/amap-maps-mcp-server"],
|
|
"env": {**os.environ, "AMAP_MAPS_API_KEY": AMAP_MAPS_API_KEY}
|
|
},
|
|
|
|
# 方式1.2: 云端MCP服务 —— Streamable HTTP模式
|
|
# "高德地图" :{
|
|
# "transport":"streamable_http",
|
|
# "url": f"https://mcp.amap.com/mcp?key={AMAP_MAPS_API_KEY}"
|
|
# },
|
|
|
|
# 方式2.1: 本地工具 —— stdio模式
|
|
# "本地天气":{
|
|
# "transport": "stdio",
|
|
# "command": "python",
|
|
# "args": ["-m", "m10_mcp_basics.s01_stdio_server"],
|
|
# "env": None
|
|
# },
|
|
|
|
# 方式2.2:本地MCP服务 —— Streamable HTTP 模式
|
|
# 注:此方法需要提前运行m10的 s02_streamable_http_server.py
|
|
# "本地天气":{
|
|
# "transport":"streamable_http",
|
|
# "url": "http://127.0.0.1:8001/mcp"
|
|
# }
|
|
}
|
|
|
|
|
|
def build_graph(available_tools):
|
|
"""构建图逻辑 (保持不变)"""
|
|
if not available_tools:
|
|
print("⚠️ 未加载任何工具")
|
|
|
|
llm = ChatOpenAI(
|
|
model="deepseek-chat",
|
|
api_key=OPENAI_API_KEY,
|
|
base_url="https://api.deepseek.com",
|
|
streaming=True
|
|
)
|
|
|
|
llm_with_tools = llm.bind_tools(available_tools) if available_tools else llm
|
|
|
|
sys_prompt = "你是一个地理位置助手，请根据用户需求调用工具查询信息。"
|
|
|
|
async def agent_node(state: MessagesState):
|
|
# 格式化消息，确保ToolMessage的content是字符串
|
|
formatted_messages = []
|
|
for msg in state["messages"]:
|
|
if isinstance(msg,ToolMessage) and not isinstance(msg.content,str):
|
|
# 将list/dict转为JSON字符串
|
|
formatted_messages.append(
|
|
ToolMessage(
|
|
content=json.dumps(msg.content,ensure_ascii=False),
|
|
tool_call_id=msg.tool_call_id
|
|
)
|
|
)
|
|
else:
|
|
formatted_messages.append(msg)
|
|
messages = [SystemMessage(content=sys_prompt)] + formatted_messages
|
|
return {"messages": [await llm_with_tools.ainvoke(messages)]}
|
|
|
|
workflow = StateGraph(MessagesState)
|
|
workflow.add_node("agent", agent_node)
|
|
|
|
if available_tools:
|
|
workflow.add_node("tools", ToolNode(available_tools))
|
|
|
|
def should_continue(state):
|
|
last_msg = state["messages"][-1]
|
|
return "tools" if last_msg.tool_calls else END
|
|
|
|
workflow.add_edge(START, "agent")
|
|
workflow.add_conditional_edges("agent", should_continue)
|
|
workflow.add_edge("tools", "agent")
|
|
else:
|
|
workflow.add_edge(START, "agent")
|
|
workflow.add_edge("agent", END)
|
|
|
|
return workflow.compile()
|
|
|
|
|
|
async def main():
|
|
print("🔌 正在初始化 MCP 客户端...")
|
|
|
|
client = MultiServerMCPClient(MCP_SERVERS)
|
|
|
|
# 显式建立连接并获取工具
|
|
# 注意：这个 client 对象会保持连接，直到脚本结束
|
|
tools = await client.get_tools()
|
|
print(f"✅ 成功加载工具: {[t.name for t in tools]}")
|
|
|
|
# 构建并运行
|
|
app = build_graph(tools)
|
|
query = "帮我查一下杭州西湖附近的酒店"
|
|
await run_agent_with_streaming(app, query)
|
|
|
|
|
|
if __name__ == "__main__":
|
|
asyncio.run(main()) |