Short circuit on repetitive answers

[nilsherzig]

Is your feature request related to a problem? Please describe.
Some models tend to produce repetitive answers if their context fills up or they simply aren't trained on doing function calls.

Describe the solution you'd like
Stop the agent chain after n times the same answer.

Describe alternatives you've considered
Forcing the use of tested models, but this seems to go against the spirit of open source / configure ability.