Add files via upload

Author: QueryType

mcp/testclient.py

@@ -0,0 +1,47 @@
+import asyncio
+
+from agents import Agent, Runner, gen_trace_id, trace
+from agents.mcp import MCPServer, MCPServerSse
+from agents.model_settings import ModelSettings
+
+# load dotenv
+from dotenv import load_dotenv
+load_dotenv()
+
+async def run(mcp_server: MCPServer):
+    agent = Agent(
+        name="Assistant",
+        instructions="Use the tools to answer the questions.",
+        mcp_servers=[mcp_server],
+        model_settings=ModelSettings(tool_choice="required"),
+        model="gpt-4o",
+    )
+
+    
+    # Test tool call
+    while True:
+        # Get user input
+        message = input("User>> ")
+        if message == "exit":
+            return "bye"
+        else:
+            #print(f"Running: {message}")
+            result = await Runner.run(starting_agent=agent, input=message)
+            print(f"gpt4o>> {result.final_output}")
+
+async def main():
+    async with MCPServerSse(
+        name="wttr.in python mcp sse server",
+        params={
+            "url": "http://10.0.0.4:7860/sse",
+        },
+        cache_tools_list=True,
+    ) as server:
+        trace_id = gen_trace_id()
+        with trace(workflow_name="wttr.in  sse Trace", trace_id=trace_id):
+            print(f"View trace: https://platform.openai.com/traces/trace?trace_id={trace_id}\n")
+            await run(server)
+
+
+if __name__ == "__main__":
+    asyncio.run(main())

mcp/testserver.py

@@ -0,0 +1,15 @@
+from mcp.server.fastmcp import FastMCP
+
+# Create an MCP server instance named "hello"
+mcp = FastMCP("hello",host="10.0.0.4", port=7860)
+
+# Define a tool called "hello" that returns a simple greeting
+@mcp.tool()
+def hello() -> str:
+    """Returns a simple greeting."""
+    return "Hello, World!"
+
+if __name__ == "__main__":
+    # Start the MCP server using stdio transport (ideal for integration with Cursor)
+    print("Starting MCP server...")
+    mcp.run(transport="sse")    

mcp/wttr_chat_client.py

@@ -0,0 +1,56 @@
+import asyncio
+
+from agents import Agent, Runner, gen_trace_id, trace
+from agents.mcp import MCPServer, MCPServerSse
+from agents.model_settings import ModelSettings
+
+# load dotenv
+from dotenv import load_dotenv
+load_dotenv()
+
+async def run(mcp_server: MCPServer):
+    travel_expert_agent = Agent(
+        name="Travel Expert Assistant",
+        instructions="You are a travel expert.",
+        model="gpt-4o",
+    )
+
+    wttr_agent = Agent(
+        name="wttr Assistant",
+        instructions="Use the tools to answer the questions.",
+        mcp_servers=[mcp_server],
+        model_settings=ModelSettings(tool_choice="required"),
+        model="gpt-4o",
+    )
+
+    triage_agent = Agent(
+        name="Triage Agent",
+        instructions="You determine which agent to use based on the user's input.",
+        handoffs=[wttr_agent, travel_expert_agent]
+    )
+    # Test tool call
+    while True:
+        # Get user input
+        message = input("User>> ")
+        if message == "exit":
+            return "bye"
+        else:
+            result = await Runner.run(starting_agent=triage_agent, input=message)
+            print(f"gpt4o>> {result.final_output}")
+
+async def main():
+    async with MCPServerSse(
+        name="wttr python mcp sse server",
+        params={
+            "url": "http://10.0.0.4:7860/sse",
+        },
+        cache_tools_list=True,
+    ) as server:
+        trace_id = gen_trace_id()
+        with trace(workflow_name="wttr  sse Trace", trace_id=trace_id):
+            print(f"View trace: https://platform.openai.com/traces/trace?trace_id={trace_id}\n")
+            await run(server)
+
+
+if __name__ == "__main__":
+    asyncio.run(main())