@wix/evalforge-types 0.68.0 → 0.70.0

package/build/index.mjs

@@ -1030,22 +1030,6 @@ var TriggerSchema = z28.object({
   metadata: TriggerMetadataSchema.optional(),
   type: z28.nativeEnum(TriggerType)
 });
-var FailureCategory = /* @__PURE__ */ ((FailureCategory2) => {
-  FailureCategory2["MISSING_FILE"] = "missing_file";
-  FailureCategory2["WRONG_CONTENT"] = "wrong_content";
-  FailureCategory2["BUILD_ERROR"] = "build_error";
-  FailureCategory2["TEST_FAILURE"] = "test_failure";
-  FailureCategory2["RUNTIME_ERROR"] = "runtime_error";
-  FailureCategory2["PERFORMANCE"] = "performance";
-  return FailureCategory2;
-})(FailureCategory || {});
-var FailureSeverity = /* @__PURE__ */ ((FailureSeverity2) => {
-  FailureSeverity2["CRITICAL"] = "critical";
-  FailureSeverity2["HIGH"] = "high";
-  FailureSeverity2["MEDIUM"] = "medium";
-  FailureSeverity2["LOW"] = "low";
-  return FailureSeverity2;
-})(FailureSeverity || {});
 var DiffLineTypeSchema = z28.enum(["added", "removed", "unchanged"]);
 var DiffLineSchema = z28.object({
   type: DiffLineTypeSchema,
@@ -1098,20 +1082,23 @@ var ExecutionTraceSchema = z28.object({
   apiCalls: z28.array(ApiCallSchema),
   totalDuration: z28.number()
 });
-var FailureAnalysisSchema = z28.object({
-  category: z28.enum(FailureCategory),
-  severity: z28.enum(FailureSeverity),
+var RunAnalysisFindingSchema = z28.object({
+  category: z28.enum([
+    "failure_pattern",
+    "cost_waste",
+    "flakiness",
+    "inefficiency",
+    "positive"
+  ]),
+  severity: z28.enum(["high", "medium", "low"]),
+  description: z28.string(),
+  affectedScenarios: z28.array(z28.string()),
+  recommendation: z28.string().optional()
+});
+var RunAnalysisSchema = z28.object({
+  generatedAt: z28.string(),
   summary: z28.string(),
-  details: z28.string(),
-  rootCause: z28.string(),
-  suggestedFix: z28.string(),
-  relatedAssertions: z28.array(z28.string()),
-  codeSnippet: z28.string().optional(),
-  similarIssues: z28.array(z28.string()).optional(),
-  patternId: z28.string().optional(),
-  // Extended fields for detailed debugging
-  diff: DiffContentSchema.optional(),
-  executionTrace: ExecutionTraceSchema.optional()
+  findings: z28.array(RunAnalysisFindingSchema)
 });
 var EvalRunSchema = TenantEntitySchema.extend({
   /** Agent ID for this run */
@@ -1132,8 +1119,6 @@ var EvalRunSchema = TenantEntitySchema.extend({
   results: z28.array(z28.lazy(() => EvalRunResultSchema)),
   /** Aggregated metrics across all results */
   aggregateMetrics: EvalMetricsSchema,
-  /** Failure analyses */
-  failureAnalyses: z28.array(FailureAnalysisSchema).optional(),
   /** Aggregated LLM trace summary */
   llmTraceSummary: LLMTraceSummarySchema.optional(),
   /** What triggered this run */
@@ -1169,7 +1154,13 @@ var EvalRunSchema = TenantEntitySchema.extend({
     runCommand: AgentRunCommandSchema.optional(),
     systemPrompt: z28.string().nullable().optional(),
     modelConfig: ModelConfigSchema.optional()
-  }).optional()
+  }).optional(),
+  /** UUID linking all runs in a comparison group */
+  comparisonGroupId: z28.string().optional(),
+  /** Human-readable label for this variant (e.g., "MCP: Wix Stores") */
+  comparisonLabel: z28.string().optional(),
+  /** LLM-generated analysis of the completed run */
+  runAnalysis: RunAnalysisSchema.optional()
 });
 var CreateEvalRunInputSchema = EvalRunSchema.omit({
   id: true,
@@ -1805,9 +1796,6 @@ export {
   EvaluationResultSchema,
   ExecutionTraceSchema,
   ExpectedFileSchema,
-  FailureAnalysisSchema,
-  FailureCategory,
-  FailureSeverity,
   FileContentCheckSchema,
   FileContentTestSchema,
   FileModificationSchema,
@@ -1844,6 +1832,8 @@ export {
   RUN_COMMAND_LABELS,
   RuleSchema,
   RuleTypeSchema,
+  RunAnalysisFindingSchema,
+  RunAnalysisSchema,
   SEMVER_REGEX,
   SKILL_FOLDER_NAME_REGEX,
   SYSTEM_ASSERTIONS,