/** * First-class eval primitives for agent quality checks. * * @module eval * * @example * ```ts * import { datasets, evalAgent, metrics } from "veryfront/eval"; * * export default evalAgent({ * target: "agent:researcher", * dataset: datasets.inline([ * { id: "q1", input: "Capital of France?", reference: "Paris" }, * ]), * metrics: [ * metrics.answer.contains({ text: "Paris" }).gate(), * metrics.agent.calledTool("search_docs").gate(), * metrics.agent.noFailedTools().gate(), * ], * }); * ``` * * @example Live agent-service eval * ```ts * import { datasets, evalAgent, metrics, runEval } from "veryfront/eval"; * import { createAgentServiceEvalAdapter } from "veryfront/eval/agent-service"; * * const definition = evalAgent({ * target: "agent:veryfront", * dataset: datasets.inline([{ id: "smoke", input: "List project files." }]), * metrics: [metrics.agent.noFailedTools().gate()], * }); * * const report = await runEval(definition, { * adapters: { * agent: createAgentServiceEvalAdapter({ * endpoint: "http://127.0.0.1:3001/api/ag-ui", * authToken: "", * projectId: "", * }), * }, * }); * ``` */ import "../../_dnt.polyfills.js"; export { datasets } from "./datasets.js"; export { evalAgent, evalTool, isEvalDefinition } from "./factory.js"; export { judges } from "./judges.js"; export { metrics } from "./metrics.js"; export { createEvalDatasetMetadata, createEvalReport, EVAL_REPORT_SCHEMA_VERSION, summarizeEvalRecords, } from "./report.js"; export { compareEvalReports } from "./baseline.js"; export { compareEvalModelReports, createEvalModelComparisonMarkdown } from "./model-comparison.js"; export { createEvalRunProvenance, resolveEvalRunProvenance } from "./provenance.js"; export { createEvalRunId } from "./run-id.js"; export { exportEvalReport, runEval } from "./runner.js"; export { deriveEvalId, discoverEvals, findEvalById } from "./discovery.js"; export { createEvalSourceDocument, getEvalEditableFieldSchema, getEvalRunSchema, getEvalSourceDocumentSchema, getEvalSourcePatchSchema, getEvalSourceReferenceSchema, getEvalStudioCapabilitySchema, getEvalTargetKindSchema, } from "./studio.js"; export type { DiscoveredEval, EvalDiscoveryOptions, EvalDiscoveryResult } from "./discovery.js"; export type { EvalLlmGroundednessJudgeOptions, EvalLlmRubricJudgeOptions } from "./judges.js"; export type { EvalAgentAdapter, EvalAgentAdapterContext, EvalAgentAdapterResult, EvalAgentInput, EvalAnswerGroundednessMetricOptions, EvalBudgetDeltaSummary, EvalCheckContext, EvalCitation, EvalDataset, EvalDatasetLoadContext, EvalDefinition, EvalDurationSummary, EvalExample, EvalExampleInput, EvalExpect, EvalExpectation, EvalFailedExampleSummary, EvalFlakeSummary, EvalGateFailureSummary, EvalKnowledgeCitationMetricOptions, EvalKnowledgeExpectedSource, EvalKnowledgeMrrMetricOptions, EvalKnowledgeRetrievalMetricOptions, EvalMetric, EvalMetricContext, EvalMetricDeltaSummary, EvalMetricFamily, EvalMetricResult, EvalMetricSummary, EvalMetricThreshold, EvalMockTools, EvalMockToolsResolver, EvalMockToolsResolverContext, EvalModelCandidateComparison, EvalModelComparison, EvalModelComparisonConstraint, EvalModelComparisonDecision, EvalModelComparisonMetricName, EvalModelComparisonObjective, EvalModelComparisonOptions, EvalModelReportSummary, EvalRecord, EvalReport, EvalReportComparison, EvalReportComparisonPolicy, EvalReportDatasetMetadata, EvalReportExportConfig, EvalReportMetadata, EvalReportSummary, EvalRetrievedContext, EvalRunProvenance, EvalSeverity, EvalSource, EvalTargetKind, EvalToolAdapter, EvalToolAdapterContext, EvalToolAdapterResult, EvalToolCall, EvalToolCallCountOptions, EvalToolCallMatchOptions, EvalToolCallStatus, EvalToolInput, EvalToolInputMatchMode, EvalTrace, EvalUsage, EvalUsageSummary, RunEvalOptions, } from "./types.js"; export type { CreateEvalSourceDocumentOptions, EvalEditableField, EvalRun, EvalSourceDocument, EvalSourcePatch, EvalSourceReference, EvalStudioCapability, } from "./studio.js"; //# sourceMappingURL=index.d.ts.map