Working example- httpstreamable #938

Activity

1. tampiss

   added

   questionFurther information is requestedFurther information is requested

   on Jun 11, 2025
2. mjunaidca commented on Jun 11, 2025

   

   mjunaidca

   on Jun 11, 2025

   More actions

   Here's a one using streamable http stateless transport in python with a simple httpx client:

   https://github.com/panaversity/learn-agentic-ai/tree/main/03_ai_protocols/02_model_context_protocol/01_server_development/01_hello_mcp_server
3. tampiss commented on Jun 12, 2025

   

   tampiss

   on Jun 12, 2025

   Author

   More actions

   Here's a one using streamable http stateless transport in python with a simple httpx client:

   https://github.com/panaversity/learn-agentic-ai/tree/main/03_ai_protocols/02_model_context_protocol/01_server_development/01_hello_mcp_server

   Thank you, but unfortunately, it is the same as with other approaches. It does not work with the clients I use due to the redirect issue, which still occurs in the Py SDK.

   INFO: 10.78.80.197:53202 - "GET /mcp HTTP/1.1" 307 Temporary Redirect
   INFO: 10.78.80.197:53202 - "GET /mcp/ HTTP/1.1" 200 OK
4. xuzexin-hz commented on Jun 20, 2025

   

   xuzexin-hz

   on Jun 20, 2025

   Contributor

   More actions

   Yes, whether printing mcp logs or checking other tools calling mcp, the /mcp/ address is ok. Although I don't know why the underlying layer does this
5. pamelafox commented on Jul 9, 2025

   

   pamelafox

   on Jul 9, 2025

   Contributor

   More actions

   This one works for me:

   """MCP Streamable HTTP Client"""
   import argparse
   import asyncio
   import logging
   
   from mcp import ClientSession
   from mcp.client.streamable_http import streamablehttp_client
   
   async def main():
       """Main function to run the MCP client"""
       parser = argparse.ArgumentParser()
       # If the server returns a redirect for /mcp to /mcp/, then just specify /mcp/ in the URL
       parser.add_argument(
           "--server", type=str, default="http://localhost:8000/mcp", help="Server hostname"
       )
       args = parser.parse_args()
   
       async with streamablehttp_client(args.server) as (
           read_stream,
           write_stream,
           _,
       ):
           # Create a session using the client streams
           async with ClientSession(read_stream, write_stream) as session:
               # Initialize the connection
               await session.initialize()
               # List available tools
               tools_list = await session.list_tools()
               print("Available tools:", ", ".join(tool.name for tool in tools_list.tools))
               # Call a tool
               tool_result = await session.call_tool("suggest_hotels", arguments={"location": "San Francisco", "check_in": "2023-10-01", "check_out": "2023-10-05"})
               print("Tool result:", tool_result.content[0].text)
   
   if __name__ == "__main__":
       logging.basicConfig(level=logging.DEBUG)
       asyncio.run(main())
   

   👍React with 👍1Reacted by Felix Weinberger
6. felixweinberger commented on Oct 7, 2025

   

   felixweinberger

   on Oct 7, 2025

   Contributor

   More actions

   See above example.
7. felixweinberger

   closed this as completedon Oct 7, 2025