Implementing sampling using python SDK for complete agentic behavior

[chaitanyadav69]

Pre-submission Checklist

• I have checked that this question would not be more appropriate as an issue in a specific repository
• I have searched existing discussions and documentation for answers

Question Category

• Protocol Specification
• SDK Usage
• Server Implementation
• General Implementation
• Documentation
• Other

Your Question

I am Building a MCP server for our hospital management application and we have implemented mcp tools such as registration, appointment and billing. Also we have incorporated speech to text. And this part works properly to allow patients to interact with our application through LLM based NLP interface. As a next step to this we are trying to implement sampling, so that our application can query back to the patient through LLM. So in this context following clarification is requested.

1. How to invoke sampling/createmessage from the application end?
2. Do I need to pass mcp server address in the createmessage method?
3. How the LLM session will be accessed through createmessage to enable LLM to do the completion?

Thanks in advance