Do we want to include a flowchart similar to the following (with appropriate cleanup/simplification) somewhere too?

flowchart LR
    B{{What should the server do?}}
    
    B -->|Provide context to LLM| C[Define resources in server]
    C --> C1[Static data]
    C --> C2[Dynamic data]
    C --> C3[Real-time updates]
    
    B -->|Enable agentic behavior| D[Define tools in server]
    D --> D1[Actions for LLM to invoke]
    D --> D2[Integrate with external systems]
    
    B -->|Define reusable LLM interactions| E[Define prompts in server]
    E --> E1[Templated prompts]
    E --> E2[Predefined LLM tasks]
    
    B -->|Custom LLM processing| F[Server sampling from client]
    F --> F1[Recursive LLM operations]
    F --> F2[Data preprocessing]

Do we want to include a flowchart similar to the following (with appropriate cleanup/simplification) somewhere too?

flowchart LR
    B{{What should the server do?}}
    
    B -->|Provide context to LLM| C[Define resources in server]
    C --> C1[Static data]
    C --> C2[Dynamic data]
    C --> C3[Real-time updates]
    
    B -->|Enable agentic behavior| D[Define tools in server]
    D --> D1[Actions for LLM to invoke]
    D --> D2[Integrate with external systems]
    
    B -->|Define reusable LLM interactions| E[Define prompts in server]
    E --> E1[Templated prompts]
    E --> E2[Predefined LLM tasks]
    
    B -->|Custom LLM processing| F[Server sampling from client]
    F --> F1[Recursive LLM operations]
    F --> F2[Data preprocessing]

Loading

I am a bit worried that unless you are super into the topic, this is more confusing than clarifying. We could add an "in-depth appendix or so", but for now I'd keep it out of the docs unless you think it would be super useful.