@wix/evalforge-types 0.66.0 → 0.68.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/build/index.js CHANGED
@@ -1368,7 +1368,17 @@ var EvalRunSchema = TenantEntitySchema.extend({
1368
1368
  /** Rule IDs to enable for this run (optional) */
1369
1369
  ruleIds: import_zod28.z.array(import_zod28.z.string()).optional(),
1370
1370
  /** Tags used to select scenarios for this run (for traceability) */
1371
- tags: import_zod28.z.array(import_zod28.z.string()).optional()
1371
+ tags: import_zod28.z.array(import_zod28.z.string()).optional(),
1372
+ /** How many times each scenario is executed within this eval run. Default: 1. Max: 20. */
1373
+ runsPerScenario: import_zod28.z.number().int().min(1).max(20).optional(),
1374
+ /** Snapshot of agent configuration captured at run creation time */
1375
+ agentSnapshot: import_zod28.z.object({
1376
+ name: import_zod28.z.string().optional(),
1377
+ agentType: AgentTypeSchema.optional(),
1378
+ runCommand: AgentRunCommandSchema.optional(),
1379
+ systemPrompt: import_zod28.z.string().nullable().optional(),
1380
+ modelConfig: ModelConfigSchema.optional()
1381
+ }).optional()
1372
1382
  });
1373
1383
  var CreateEvalRunInputSchema = EvalRunSchema.omit({
1374
1384
  id: true,
@@ -1380,7 +1390,8 @@ var CreateEvalRunInputSchema = EvalRunSchema.omit({
1380
1390
  aggregateMetrics: true,
1381
1391
  startedAt: true,
1382
1392
  completedAt: true,
1383
- scenarioIds: true
1393
+ scenarioIds: true,
1394
+ agentSnapshot: true
1384
1395
  }).extend({
1385
1396
  /** Optional on input — backend resolves from tags when not provided */
1386
1397
  scenarioIds: import_zod28.z.array(import_zod28.z.string()).optional()
@@ -1492,6 +1503,8 @@ var EvalRunResultSchema = import_zod30.z.object({
1492
1503
  skillVersion: import_zod30.z.string().optional(),
1493
1504
  scenarioId: import_zod30.z.string(),
1494
1505
  scenarioName: import_zod30.z.string(),
1506
+ /** Snapshot of the trigger prompt used during the run (prevents stale display after edits) */
1507
+ triggerPrompt: import_zod30.z.string().optional(),
1495
1508
  modelConfig: ModelConfigSchema.optional(),
1496
1509
  assertionResults: import_zod30.z.array(AssertionResultSchema),
1497
1510
  metrics: EvalMetricsSchema.optional(),
@@ -1508,7 +1521,9 @@ var EvalRunResultSchema = import_zod30.z.object({
1508
1521
  completedAt: import_zod30.z.string().optional(),
1509
1522
  llmTrace: LLMTraceSchema.optional(),
1510
1523
  /** Full conversation messages (only present in transit; stripped before DB storage) */
1511
- conversation: import_zod30.z.array(ConversationMessageSchema).optional()
1524
+ conversation: import_zod30.z.array(ConversationMessageSchema).optional(),
1525
+ /** 0-based iteration index when a scenario is run multiple times within a single eval run */
1526
+ iterationIndex: import_zod30.z.number().int().min(0).optional()
1512
1527
  });
1513
1528
  var PromptResultSchema = import_zod30.z.object({
1514
1529
  text: import_zod30.z.string(),