Agent/Intra-node Learning Example

[hinthornw]

Similar to GPTSwarm: https://github.com/hinthornw/GPTSwarm

Initially, skip the edge optimization flows (cool as they are) - simple example of node optimization.

Likely via few-shot examples, though other optimizers could be applied