Writing a Genkit Evaluator | Firebase
I have started looking into evaluators, it's been hiding from me somehow and looks cool, however, it's difficult to wrap my head around it.
For start, I would like to test whether my flow is still doing what it should after any changes, so I would like to just verify that input -> output = expected output
Closest to this I could fine was testing ending of custom evaluators, which had a file like that: https://firebase.google.com/docs/genkit/plugin-authoring-evaluator#testing
Is that a way I should go please? Writing a custom evaluator that does just equality check? (or probably something like deep equal for the structured output)
For start, I would like to test whether my flow is still doing what it should after any changes, so I would like to just verify that input -> output = expected output
Closest to this I could fine was testing ending of custom evaluators, which had a file like that: https://firebase.google.com/docs/genkit/plugin-authoring-evaluator#testing
Is that a way I should go please? Writing a custom evaluator that does just equality check? (or probably something like deep equal for the structured output)
