Is your feature request related to a problem? Please describe.
MCP protocol can send prompts to clients, however, prompts are often tuned for certain models or types of models.
E.g. reasoning models, coding models, etc.
I didn't see a way to send metadata along with the prompt from server to client.
Describe the solution you'd like
If there was a way to attach metadata to the prompts using attributes using
name/value pairs that would be returned with the prompt to the client that would be helpful.
e.g. [Meta(Name="Model", Value="gpt-4o")]
Or some other.
Describe alternatives you've considered
There may be better ways to do it.
Additional context
We spend a lot of time tuning prompts against different models, so having alternative versions of the prompt and being able to tag or sort them out based on the client-side model availability would be helpful.
Is your feature request related to a problem? Please describe.
MCP protocol can send prompts to clients, however, prompts are often tuned for certain models or types of models.
E.g. reasoning models, coding models, etc.
I didn't see a way to send metadata along with the prompt from server to client.
Describe the solution you'd like
If there was a way to attach metadata to the prompts using attributes using
name/value pairs that would be returned with the prompt to the client that would be helpful.
e.g.
[Meta(Name="Model", Value="gpt-4o")]Or some other.
Describe alternatives you've considered
There may be better ways to do it.
Additional context
We spend a lot of time tuning prompts against different models, so having alternative versions of the prompt and being able to tag or sort them out based on the client-side model availability would be helpful.