@wix/evalforge-types 0.68.0 → 0.70.0

package/build/index.js

@@ -100,9 +100,6 @@ __export(index_exports, {
   EvaluationResultSchema: () => EvaluationResultSchema,
   ExecutionTraceSchema: () => ExecutionTraceSchema,
   ExpectedFileSchema: () => ExpectedFileSchema,
-  FailureAnalysisSchema: () => FailureAnalysisSchema,
-  FailureCategory: () => FailureCategory,
-  FailureSeverity: () => FailureSeverity,
   FileContentCheckSchema: () => FileContentCheckSchema,
   FileContentTestSchema: () => FileContentTestSchema,
   FileModificationSchema: () => FileModificationSchema,
@@ -139,6 +136,8 @@ __export(index_exports, {
   RUN_COMMAND_LABELS: () => RUN_COMMAND_LABELS,
   RuleSchema: () => RuleSchema,
   RuleTypeSchema: () => RuleTypeSchema,
+  RunAnalysisFindingSchema: () => RunAnalysisFindingSchema,
+  RunAnalysisSchema: () => RunAnalysisSchema,
   SEMVER_REGEX: () => SEMVER_REGEX,
   SKILL_FOLDER_NAME_REGEX: () => SKILL_FOLDER_NAME_REGEX,
   SYSTEM_ASSERTIONS: () => SYSTEM_ASSERTIONS,
@@ -1239,22 +1238,6 @@ var TriggerSchema = import_zod28.z.object({
   metadata: TriggerMetadataSchema.optional(),
   type: import_zod28.z.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 = import_zod28.z.enum(["added", "removed", "unchanged"]);
 var DiffLineSchema = import_zod28.z.object({
   type: DiffLineTypeSchema,
@@ -1307,20 +1290,23 @@ var ExecutionTraceSchema = import_zod28.z.object({
   apiCalls: import_zod28.z.array(ApiCallSchema),
   totalDuration: import_zod28.z.number()
 });
-var FailureAnalysisSchema = import_zod28.z.object({
-  category: import_zod28.z.enum(FailureCategory),
-  severity: import_zod28.z.enum(FailureSeverity),
+var RunAnalysisFindingSchema = import_zod28.z.object({
+  category: import_zod28.z.enum([
+    "failure_pattern",
+    "cost_waste",
+    "flakiness",
+    "inefficiency",
+    "positive"
+  ]),
+  severity: import_zod28.z.enum(["high", "medium", "low"]),
+  description: import_zod28.z.string(),
+  affectedScenarios: import_zod28.z.array(import_zod28.z.string()),
+  recommendation: import_zod28.z.string().optional()
+});
+var RunAnalysisSchema = import_zod28.z.object({
+  generatedAt: import_zod28.z.string(),
   summary: import_zod28.z.string(),
-  details: import_zod28.z.string(),
-  rootCause: import_zod28.z.string(),
-  suggestedFix: import_zod28.z.string(),
-  relatedAssertions: import_zod28.z.array(import_zod28.z.string()),
-  codeSnippet: import_zod28.z.string().optional(),
-  similarIssues: import_zod28.z.array(import_zod28.z.string()).optional(),
-  patternId: import_zod28.z.string().optional(),
-  // Extended fields for detailed debugging
-  diff: DiffContentSchema.optional(),
-  executionTrace: ExecutionTraceSchema.optional()
+  findings: import_zod28.z.array(RunAnalysisFindingSchema)
 });
 var EvalRunSchema = TenantEntitySchema.extend({
   /** Agent ID for this run */
@@ -1341,8 +1327,6 @@ var EvalRunSchema = TenantEntitySchema.extend({
   results: import_zod28.z.array(import_zod28.z.lazy(() => EvalRunResultSchema)),
   /** Aggregated metrics across all results */
   aggregateMetrics: EvalMetricsSchema,
-  /** Failure analyses */
-  failureAnalyses: import_zod28.z.array(FailureAnalysisSchema).optional(),
   /** Aggregated LLM trace summary */
   llmTraceSummary: LLMTraceSummarySchema.optional(),
   /** What triggered this run */
@@ -1378,7 +1362,13 @@ var EvalRunSchema = TenantEntitySchema.extend({
     runCommand: AgentRunCommandSchema.optional(),
     systemPrompt: import_zod28.z.string().nullable().optional(),
     modelConfig: ModelConfigSchema.optional()
-  }).optional()
+  }).optional(),
+  /** UUID linking all runs in a comparison group */
+  comparisonGroupId: import_zod28.z.string().optional(),
+  /** Human-readable label for this variant (e.g., "MCP: Wix Stores") */
+  comparisonLabel: import_zod28.z.string().optional(),
+  /** LLM-generated analysis of the completed run */
+  runAnalysis: RunAnalysisSchema.optional()
 });
 var CreateEvalRunInputSchema = EvalRunSchema.omit({
   id: true,
@@ -2015,9 +2005,6 @@ function getSystemAssertion(id) {
   EvaluationResultSchema,
   ExecutionTraceSchema,
   ExpectedFileSchema,
-  FailureAnalysisSchema,
-  FailureCategory,
-  FailureSeverity,
   FileContentCheckSchema,
   FileContentTestSchema,
   FileModificationSchema,
@@ -2054,6 +2041,8 @@ function getSystemAssertion(id) {
   RUN_COMMAND_LABELS,
   RuleSchema,
   RuleTypeSchema,
+  RunAnalysisFindingSchema,
+  RunAnalysisSchema,
   SEMVER_REGEX,
   SKILL_FOLDER_NAME_REGEX,
   SYSTEM_ASSERTIONS,