Replies: 1 comment
-
I think this is outside the scope of MCP discussion, as it relates specifically to Claude and its training. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Pre-submission Checklist
Discussion Topic
I have followed MCP client and MCP server weather example.
What I have noticed is that Claude seems over thinking and try to decide what is 'today'
Due to timezone difference, is 12/11/2024 as time of requesting for Sydney.
This is not 100% reproducible, I get the error about 50% of time. My question is, if claude is a pretrained model with a fix cut off date, what is happening here? does it do something more? and how can I make Claude understand timezone?
Thanks
I added more logs, and it shows MCP server is returning correct results.
Beta Was this translation helpful? Give feedback.
All reactions