🪢 Open source agent evals & observability: Trace, evaluate, and improve LLM applications with one open platform.
TypeScript analyticsautogenevaluationlangchainlarge-language-modelsllama-indexllmllm-evaluationllm-observabilityllmopsmonitoringobservabilityopen-sourceopenaiplaygroundprompt-engineeringprompt-managementself-hostedycombinator
import preview from "../../../../../.storybook/preview";
import { fn } from "storybook/test";
import type { InAppAgentQuestionnaireQuestion } from "@langfuse/shared/in-app-agent";
import { InAppAgentQuestionnaireCard } from "./InAppAgentQuestionnaireCard";
const environmentQuestion: InAppAgentQuestionnaireQuestion = {
question: "Which environment should the evaluator run on?",
options: [
{
label: "Production",
description: "Score live traffic as it arrives.",
recommended: true,
},
{
label: "Staging",
description: "Try it on pre-release traces first.",
recommended: false,
},
],
allowMultiple: false,
};
const failureSignalsQuestion: InAppAgentQuestionnaireQuestion = {
question: "Which signals should count as a failure?",
options: [
{
label: "Tool errors",
description: "Any tool call that returned an error.",
recommended: true,
},
{
label: "Empty output",
description: "The model returned no text.",
recommended: false,
},
{
label: "High latency",
description: "The trace took longer than 30 seconds.",
recommended: false,
},
],
allowMultiple: true,
};
const scheduleQuestion: InAppAgentQuestionnaireQuestion = {
question: "How often should it run?",
options: [
{
label: "Every hour",
description: "Catches regressions fastest.",
recommended: false,
},
{
label: "Every 6 hours",
description: "A few runs per working day.",
recommended: false,
},
{
label: "Once a day",
description: "Enough for slow-moving traffic.",
recommended: false,
},
{
label: "Once a week",
description: "Lowest cost, slowest signal.",
recommended: false,
},
],
allowMultiple: false,
};
const multiStepQuestions = [
environmentQuestion,
failureSignalsQuestion,
scheduleQuestion,
];
const meta = preview.meta({
component: InAppAgentQuestionnaireCard,
args: {
questions: [environmentQuestion],
isSubmitting: false,
onSubmit: fn().mockResolvedValue(undefined),
},
});
export const SingleChoice = meta.story({});
export const MultipleChoice = meta.story({
args: { questions: [failureSignalsQuestion] },
});
export const MultiStep = meta.story({
args: { questions: multiStepQuestions },
});