@wix/evalforge-types 0.52.0 → 0.54.0

package/build/index.js

@@ -46,6 +46,9 @@ __export(index_exports, {
   BuildCheckTestSchema: () => BuildCheckTestSchema,
   BuildPassedAssertionSchema: () => BuildPassedAssertionSchema,
   BuildPassedConfigSchema: () => BuildPassedConfigSchema,
+  BulkImportResultItemSchema: () => BulkImportResultItemSchema,
+  BulkImportResultSchema: () => BulkImportResultSchema,
+  BulkImportSkillsInputSchema: () => BulkImportSkillsInputSchema,
   ClaudeModel: () => ClaudeModel2,
   ClaudeModelSchema: () => ClaudeModelSchema,
   CommandExecutionSchema: () => CommandExecutionSchema,
@@ -1359,6 +1362,21 @@ var UpdateSkillInputSchema = SkillInputBaseSchema.partial().refine(
 var SkillWithLatestVersionSchema = SkillSchema.extend({
   latestVersion: SkillVersionSchema.optional()
 });
+var BulkImportSkillsInputSchema = import_zod7.z.object({
+  source: GitHubSourceSchema
+});
+var BulkImportResultItemSchema = import_zod7.z.object({
+  name: import_zod7.z.string(),
+  status: import_zod7.z.enum(["created", "skipped", "failed"]),
+  skillId: import_zod7.z.string().optional(),
+  reason: import_zod7.z.string().optional()
+});
+var BulkImportResultSchema = import_zod7.z.object({
+  created: import_zod7.z.number(),
+  skipped: import_zod7.z.number(),
+  failed: import_zod7.z.number(),
+  items: import_zod7.z.array(BulkImportResultItemSchema)
+});
 
 // src/target/sub-agent.ts
 var import_zod8 = require("zod");
@@ -1591,85 +1609,34 @@ var TestSchema = import_zod20.z.discriminatedUnion("type", [
   PlaywrightNLTestSchema
 ]);
 
-// src/scenario/assertions.ts
-var import_zod21 = require("zod");
-var SkillWasCalledAssertionSchema = import_zod21.z.object({
-  type: import_zod21.z.literal("skill_was_called"),
-  /** Names of the skills that must have been called (matched against trace Skill tool args) */
-  skillNames: import_zod21.z.array(import_zod21.z.string().min(1)).min(1)
-});
-var ToolCalledWithParamAssertionSchema = import_zod21.z.object({
-  type: import_zod21.z.literal("tool_called_with_param"),
-  /** Name of the tool that must have been called */
-  toolName: import_zod21.z.string().min(1),
-  /** JSON string of key-value pairs for expected parameters (substring match) */
-  expectedParams: import_zod21.z.string().min(1)
-});
-var BuildPassedAssertionSchema = import_zod21.z.object({
-  type: import_zod21.z.literal("build_passed"),
-  /** Command to run (default: "yarn build") */
-  command: import_zod21.z.string().optional(),
-  /** Expected exit code (default: 0) */
-  expectedExitCode: import_zod21.z.number().int().optional()
-});
-var CostAssertionSchema = import_zod21.z.object({
-  type: import_zod21.z.literal("cost"),
-  /** Maximum allowed cost in USD */
-  maxCostUsd: import_zod21.z.number().positive()
-});
-var LlmJudgeAssertionSchema = import_zod21.z.object({
-  type: import_zod21.z.literal("llm_judge"),
-  /** Prompt template; placeholders: {{output}}, {{cwd}}, {{changedFiles}}, {{trace}} */
-  prompt: import_zod21.z.string(),
-  /** Minimum score to pass (0-10, default 7) */
-  minScore: import_zod21.z.number().int().min(0).max(10).optional(),
-  /** Model for the judge (e.g. claude-3-5-haiku) */
-  model: import_zod21.z.string().optional(),
-  maxTokens: import_zod21.z.number().int().optional(),
-  temperature: import_zod21.z.number().min(0).max(1).optional()
-});
-var TimeAssertionSchema = import_zod21.z.object({
-  type: import_zod21.z.literal("time_limit"),
-  /** Maximum allowed duration in milliseconds */
-  maxDurationMs: import_zod21.z.number().int().positive()
-});
-var AssertionSchema = import_zod21.z.union([
-  SkillWasCalledAssertionSchema,
-  ToolCalledWithParamAssertionSchema,
-  BuildPassedAssertionSchema,
-  TimeAssertionSchema,
-  CostAssertionSchema,
-  LlmJudgeAssertionSchema
-]);
-
 // src/scenario/environment.ts
-var import_zod22 = require("zod");
-var LocalProjectConfigSchema = import_zod22.z.object({
+var import_zod21 = require("zod");
+var LocalProjectConfigSchema = import_zod21.z.object({
   /** Template ID to use for the local project */
-  templateId: import_zod22.z.string().optional(),
+  templateId: import_zod21.z.string().optional(),
   /** Files to create in the project */
-  files: import_zod22.z.array(
-    import_zod22.z.object({
-      path: import_zod22.z.string().min(1),
-      content: import_zod22.z.string().min(1)
+  files: import_zod21.z.array(
+    import_zod21.z.object({
+      path: import_zod21.z.string().min(1),
+      content: import_zod21.z.string().min(1)
     })
   ).optional()
 });
-var MetaSiteConfigSchema = import_zod22.z.object({
-  configurations: import_zod22.z.array(
-    import_zod22.z.object({
-      name: import_zod22.z.string().min(1),
-      apiCalls: import_zod22.z.array(
-        import_zod22.z.object({
-          url: import_zod22.z.string().url(),
-          method: import_zod22.z.enum(["POST", "PUT"]),
-          body: import_zod22.z.string()
+var MetaSiteConfigSchema = import_zod21.z.object({
+  configurations: import_zod21.z.array(
+    import_zod21.z.object({
+      name: import_zod21.z.string().min(1),
+      apiCalls: import_zod21.z.array(
+        import_zod21.z.object({
+          url: import_zod21.z.string().url(),
+          method: import_zod21.z.enum(["POST", "PUT"]),
+          body: import_zod21.z.string()
         })
       )
     })
   ).optional()
 });
-var EnvironmentSchema = import_zod22.z.object({
+var EnvironmentSchema = import_zod21.z.object({
   /** Local project configuration */
   localProject: LocalProjectConfigSchema.optional(),
   /** Meta site configuration */
@@ -1677,11 +1644,11 @@ var EnvironmentSchema = import_zod22.z.object({
 });
 
 // src/scenario/test-scenario.ts
-var import_zod24 = require("zod");
+var import_zod23 = require("zod");
 
 // src/assertion/assertion.ts
-var import_zod23 = require("zod");
-var AssertionTypeSchema = import_zod23.z.enum([
+var import_zod22 = require("zod");
+var AssertionTypeSchema = import_zod22.z.enum([
   "skill_was_called",
   "tool_called_with_param",
   "build_passed",
@@ -1689,59 +1656,61 @@ var AssertionTypeSchema = import_zod23.z.enum([
   "cost",
   "llm_judge"
 ]);
-var AssertionParameterTypeSchema = import_zod23.z.enum([
+var AssertionParameterTypeSchema = import_zod22.z.enum([
   "string",
   "number",
   "boolean"
 ]);
-var AssertionParameterSchema = import_zod23.z.object({
+var AssertionParameterSchema = import_zod22.z.object({
   /** Parameter name (used as key in params object) */
-  name: import_zod23.z.string().min(1),
+  name: import_zod22.z.string().min(1),
   /** Display label for the parameter */
-  label: import_zod23.z.string().min(1),
+  label: import_zod22.z.string().min(1),
   /** Parameter type */
   type: AssertionParameterTypeSchema,
   /** Whether this parameter is required */
-  required: import_zod23.z.boolean(),
+  required: import_zod22.z.boolean(),
   /** Default value (optional, used when not provided) */
-  defaultValue: import_zod23.z.union([import_zod23.z.string(), import_zod23.z.number(), import_zod23.z.boolean()]).optional(),
+  defaultValue: import_zod22.z.union([import_zod22.z.string(), import_zod22.z.number(), import_zod22.z.boolean()]).optional(),
   /** If true, parameter is hidden by default behind "Show advanced options" */
-  advanced: import_zod23.z.boolean().optional()
+  advanced: import_zod22.z.boolean().optional()
 });
-var ScenarioAssertionLinkSchema = import_zod23.z.object({
+var ScenarioAssertionLinkSchema = import_zod22.z.object({
   /** ID of the assertion (can be system assertion like 'system:skill_was_called' or custom assertion UUID) */
-  assertionId: import_zod23.z.string(),
+  assertionId: import_zod22.z.string(),
   /** Parameter values for this assertion in this scenario */
-  params: import_zod23.z.record(
-    import_zod23.z.string(),
-    import_zod23.z.union([import_zod23.z.string(), import_zod23.z.number(), import_zod23.z.boolean(), import_zod23.z.null()])
+  params: import_zod22.z.record(
+    import_zod22.z.string(),
+    import_zod22.z.union([import_zod22.z.string(), import_zod22.z.number(), import_zod22.z.boolean(), import_zod22.z.null()])
   ).optional()
 });
-var SkillWasCalledConfigSchema = import_zod23.z.object({
+var SkillWasCalledConfigSchema = import_zod22.z.object({
   /** Names of the skills that must have been called */
-  skillNames: import_zod23.z.array(import_zod23.z.string().min(1)).min(1)
+  skillNames: import_zod22.z.array(import_zod22.z.string().min(1)).min(1)
 });
-var CostConfigSchema = import_zod23.z.strictObject({
+var CostConfigSchema = import_zod22.z.strictObject({
   /** Maximum allowed cost in USD */
-  maxCostUsd: import_zod23.z.number().positive()
+  maxCostUsd: import_zod22.z.number().positive()
 });
-var ToolCalledWithParamConfigSchema = import_zod23.z.strictObject({
+var ToolCalledWithParamConfigSchema = import_zod22.z.strictObject({
   /** Name of the tool that must have been called */
-  toolName: import_zod23.z.string().min(1),
+  toolName: import_zod22.z.string().min(1),
   /** JSON string of key-value pairs for expected parameters (substring match) */
-  expectedParams: import_zod23.z.string().min(1)
+  expectedParams: import_zod22.z.string().min(1),
+  /** If true, the matching tool call must also have succeeded (step.success === true) */
+  requireSuccess: import_zod22.z.boolean().optional()
 });
-var BuildPassedConfigSchema = import_zod23.z.strictObject({
+var BuildPassedConfigSchema = import_zod22.z.strictObject({
   /** Command to run (default: "yarn build") */
-  command: import_zod23.z.string().optional(),
+  command: import_zod22.z.string().optional(),
   /** Expected exit code (default: 0) */
-  expectedExitCode: import_zod23.z.number().int().optional()
+  expectedExitCode: import_zod22.z.number().int().optional()
 });
-var TimeConfigSchema = import_zod23.z.strictObject({
+var TimeConfigSchema = import_zod22.z.strictObject({
   /** Maximum allowed duration in milliseconds */
-  maxDurationMs: import_zod23.z.number().int().positive()
+  maxDurationMs: import_zod22.z.number().int().positive()
 });
-var LlmJudgeConfigSchema = import_zod23.z.object({
+var LlmJudgeConfigSchema = import_zod22.z.object({
   /**
    * Prompt template with placeholders:
    * - {{output}}: agent's final output
@@ -1752,19 +1721,45 @@ var LlmJudgeConfigSchema = import_zod23.z.object({
    * - {{trace}}: step-by-step trace of tool calls
    * - Custom parameters defined in the parameters array
    */
-  prompt: import_zod23.z.string().min(1),
+  prompt: import_zod22.z.string().min(1),
   /** Minimum score to pass (0-10, default 7) */
-  minScore: import_zod23.z.number().int().min(0).max(10).optional(),
+  minScore: import_zod22.z.number().int().min(0).max(10).optional(),
   /** Model for the judge (e.g. claude-3-5-haiku-20241022) */
-  model: import_zod23.z.string().optional(),
+  model: import_zod22.z.string().optional(),
   /** Max output tokens */
-  maxTokens: import_zod23.z.number().int().optional(),
+  maxTokens: import_zod22.z.number().int().optional(),
   /** Temperature (0-1) */
-  temperature: import_zod23.z.number().min(0).max(1).optional(),
+  temperature: import_zod22.z.number().min(0).max(1).optional(),
   /** User-defined parameters for this assertion */
-  parameters: import_zod23.z.array(AssertionParameterSchema).optional()
+  parameters: import_zod22.z.array(AssertionParameterSchema).optional()
+});
+var SkillWasCalledAssertionSchema = SkillWasCalledConfigSchema.extend({
+  type: import_zod22.z.literal("skill_was_called")
+});
+var ToolCalledWithParamAssertionSchema = ToolCalledWithParamConfigSchema.extend({
+  type: import_zod22.z.literal("tool_called_with_param")
 });
-var AssertionConfigSchema = import_zod23.z.union([
+var BuildPassedAssertionSchema = BuildPassedConfigSchema.extend({
+  type: import_zod22.z.literal("build_passed")
+});
+var CostAssertionSchema = CostConfigSchema.extend({
+  type: import_zod22.z.literal("cost")
+});
+var LlmJudgeAssertionSchema = LlmJudgeConfigSchema.extend({
+  type: import_zod22.z.literal("llm_judge")
+});
+var TimeAssertionSchema = TimeConfigSchema.extend({
+  type: import_zod22.z.literal("time_limit")
+});
+var AssertionSchema = import_zod22.z.union([
+  SkillWasCalledAssertionSchema,
+  ToolCalledWithParamAssertionSchema,
+  BuildPassedAssertionSchema,
+  TimeAssertionSchema,
+  CostAssertionSchema,
+  LlmJudgeAssertionSchema
+]);
+var AssertionConfigSchema = import_zod22.z.union([
   LlmJudgeConfigSchema,
   // requires prompt - check first
   SkillWasCalledConfigSchema,
@@ -1777,7 +1772,7 @@ var AssertionConfigSchema = import_zod23.z.union([
   // requires maxCostUsd, uses strictObject
   BuildPassedConfigSchema,
   // all optional, uses strictObject to reject unknown keys
-  import_zod23.z.object({})
+  import_zod22.z.object({})
   // fallback empty config
 ]);
 var CustomAssertionSchema = TenantEntitySchema.extend({
@@ -1828,25 +1823,25 @@ function getLlmJudgeConfig(assertion) {
 }
 
 // src/scenario/test-scenario.ts
-var ExpectedFileSchema = import_zod24.z.object({
+var ExpectedFileSchema = import_zod23.z.object({
   /** Relative path where the file should be created */
-  path: import_zod24.z.string(),
+  path: import_zod23.z.string(),
   /** Optional expected content */
-  content: import_zod24.z.string().optional()
+  content: import_zod23.z.string().optional()
 });
 var TestScenarioSchema = TenantEntitySchema.extend({
   /** The prompt sent to the agent to trigger the task */
-  triggerPrompt: import_zod24.z.string().min(10),
+  triggerPrompt: import_zod23.z.string().min(10),
   /** ID of the template to use for this scenario (null = no template) */
-  templateId: import_zod24.z.string().nullish(),
+  templateId: import_zod23.z.string().nullish(),
   /** Inline assertions to evaluate for this scenario (legacy) */
-  assertions: import_zod24.z.array(AssertionSchema).optional(),
+  assertions: import_zod23.z.array(AssertionSchema).optional(),
   /** IDs of saved assertions to evaluate (from assertions table) - legacy, use assertionLinks */
-  assertionIds: import_zod24.z.array(import_zod24.z.string()).optional(),
+  assertionIds: import_zod23.z.array(import_zod23.z.string()).optional(),
   /** Linked assertions with per-scenario parameter values */
-  assertionLinks: import_zod24.z.array(ScenarioAssertionLinkSchema).optional(),
+  assertionLinks: import_zod23.z.array(ScenarioAssertionLinkSchema).optional(),
   /** Tags for categorisation and filtering */
-  tags: import_zod24.z.array(import_zod24.z.string()).optional()
+  tags: import_zod23.z.array(import_zod23.z.string()).optional()
 });
 var CreateTestScenarioInputSchema = TestScenarioSchema.omit({
   id: true,
@@ -1857,10 +1852,10 @@ var CreateTestScenarioInputSchema = TestScenarioSchema.omit({
 var UpdateTestScenarioInputSchema = CreateTestScenarioInputSchema.partial();
 
 // src/suite/test-suite.ts
-var import_zod25 = require("zod");
+var import_zod24 = require("zod");
 var TestSuiteSchema = TenantEntitySchema.extend({
   /** IDs of test scenarios in this suite */
-  scenarioIds: import_zod25.z.array(import_zod25.z.string())
+  scenarioIds: import_zod24.z.array(import_zod24.z.string())
 });
 var CreateTestSuiteInputSchema = TestSuiteSchema.omit({
   id: true,
@@ -1871,21 +1866,21 @@ var CreateTestSuiteInputSchema = TestSuiteSchema.omit({
 var UpdateTestSuiteInputSchema = CreateTestSuiteInputSchema.partial();
 
 // src/evaluation/metrics.ts
-var import_zod26 = require("zod");
-var TokenUsageSchema = import_zod26.z.object({
-  prompt: import_zod26.z.number(),
-  completion: import_zod26.z.number(),
-  total: import_zod26.z.number()
-});
-var EvalMetricsSchema = import_zod26.z.object({
-  totalAssertions: import_zod26.z.number(),
-  passed: import_zod26.z.number(),
-  failed: import_zod26.z.number(),
-  skipped: import_zod26.z.number(),
-  errors: import_zod26.z.number(),
-  passRate: import_zod26.z.number(),
-  avgDuration: import_zod26.z.number(),
-  totalDuration: import_zod26.z.number()
+var import_zod25 = require("zod");
+var TokenUsageSchema = import_zod25.z.object({
+  prompt: import_zod25.z.number(),
+  completion: import_zod25.z.number(),
+  total: import_zod25.z.number()
+});
+var EvalMetricsSchema = import_zod25.z.object({
+  totalAssertions: import_zod25.z.number(),
+  passed: import_zod25.z.number(),
+  failed: import_zod25.z.number(),
+  skipped: import_zod25.z.number(),
+  errors: import_zod25.z.number(),
+  passRate: import_zod25.z.number(),
+  avgDuration: import_zod25.z.number(),
+  totalDuration: import_zod25.z.number()
 });
 var EvalStatus = /* @__PURE__ */ ((EvalStatus2) => {
   EvalStatus2["PENDING"] = "pending";
@@ -1895,7 +1890,7 @@ var EvalStatus = /* @__PURE__ */ ((EvalStatus2) => {
   EvalStatus2["CANCELLED"] = "cancelled";
   return EvalStatus2;
 })(EvalStatus || {});
-var EvalStatusSchema = import_zod26.z.enum(EvalStatus);
+var EvalStatusSchema = import_zod25.z.enum(EvalStatus);
 var LLMStepType = /* @__PURE__ */ ((LLMStepType2) => {
   LLMStepType2["COMPLETION"] = "completion";
   LLMStepType2["TOOL_USE"] = "tool_use";
@@ -1903,52 +1898,52 @@ var LLMStepType = /* @__PURE__ */ ((LLMStepType2) => {
   LLMStepType2["THINKING"] = "thinking";
   return LLMStepType2;
 })(LLMStepType || {});
-var LLMTraceStepSchema = import_zod26.z.object({
-  id: import_zod26.z.string(),
-  stepNumber: import_zod26.z.number(),
-  type: import_zod26.z.enum(LLMStepType),
-  model: import_zod26.z.string(),
-  provider: import_zod26.z.string(),
-  startedAt: import_zod26.z.string(),
-  durationMs: import_zod26.z.number(),
+var LLMTraceStepSchema = import_zod25.z.object({
+  id: import_zod25.z.string(),
+  stepNumber: import_zod25.z.number(),
+  type: import_zod25.z.enum(LLMStepType),
+  model: import_zod25.z.string(),
+  provider: import_zod25.z.string(),
+  startedAt: import_zod25.z.string(),
+  durationMs: import_zod25.z.number(),
   tokenUsage: TokenUsageSchema,
-  costUsd: import_zod26.z.number(),
-  toolName: import_zod26.z.string().optional(),
-  toolArguments: import_zod26.z.string().optional(),
-  inputPreview: import_zod26.z.string().optional(),
-  outputPreview: import_zod26.z.string().optional(),
-  success: import_zod26.z.boolean(),
-  error: import_zod26.z.string().optional()
-});
-var LLMBreakdownStatsSchema = import_zod26.z.object({
-  count: import_zod26.z.number(),
-  durationMs: import_zod26.z.number(),
-  tokens: import_zod26.z.number(),
-  costUsd: import_zod26.z.number()
-});
-var LLMTraceSummarySchema = import_zod26.z.object({
-  totalSteps: import_zod26.z.number(),
-  totalDurationMs: import_zod26.z.number(),
+  costUsd: import_zod25.z.number(),
+  toolName: import_zod25.z.string().optional(),
+  toolArguments: import_zod25.z.string().optional(),
+  inputPreview: import_zod25.z.string().optional(),
+  outputPreview: import_zod25.z.string().optional(),
+  success: import_zod25.z.boolean(),
+  error: import_zod25.z.string().optional()
+});
+var LLMBreakdownStatsSchema = import_zod25.z.object({
+  count: import_zod25.z.number(),
+  durationMs: import_zod25.z.number(),
+  tokens: import_zod25.z.number(),
+  costUsd: import_zod25.z.number()
+});
+var LLMTraceSummarySchema = import_zod25.z.object({
+  totalSteps: import_zod25.z.number(),
+  totalDurationMs: import_zod25.z.number(),
   totalTokens: TokenUsageSchema,
-  totalCostUsd: import_zod26.z.number(),
-  stepTypeBreakdown: import_zod26.z.record(import_zod26.z.string(), LLMBreakdownStatsSchema).optional(),
-  modelBreakdown: import_zod26.z.record(import_zod26.z.string(), LLMBreakdownStatsSchema),
-  modelsUsed: import_zod26.z.array(import_zod26.z.string())
-});
-var LLMTraceSchema = import_zod26.z.object({
-  id: import_zod26.z.string(),
-  steps: import_zod26.z.array(LLMTraceStepSchema),
+  totalCostUsd: import_zod25.z.number(),
+  stepTypeBreakdown: import_zod25.z.record(import_zod25.z.string(), LLMBreakdownStatsSchema).optional(),
+  modelBreakdown: import_zod25.z.record(import_zod25.z.string(), LLMBreakdownStatsSchema),
+  modelsUsed: import_zod25.z.array(import_zod25.z.string())
+});
+var LLMTraceSchema = import_zod25.z.object({
+  id: import_zod25.z.string(),
+  steps: import_zod25.z.array(LLMTraceStepSchema),
   summary: LLMTraceSummarySchema
 });
 
 // src/evaluation/eval-result.ts
-var import_zod30 = require("zod");
+var import_zod29 = require("zod");
 
 // src/evaluation/eval-run.ts
-var import_zod28 = require("zod");
+var import_zod27 = require("zod");
 
 // src/evaluation/live-trace.ts
-var import_zod27 = require("zod");
+var import_zod26 = require("zod");
 var LiveTraceEventType = /* @__PURE__ */ ((LiveTraceEventType2) => {
   LiveTraceEventType2["THINKING"] = "thinking";
   LiveTraceEventType2["TOOL_USE"] = "tool_use";
@@ -1962,37 +1957,37 @@ var LiveTraceEventType = /* @__PURE__ */ ((LiveTraceEventType2) => {
   LiveTraceEventType2["USER"] = "user";
   return LiveTraceEventType2;
 })(LiveTraceEventType || {});
-var LiveTraceEventSchema = import_zod27.z.object({
+var LiveTraceEventSchema = import_zod26.z.object({
   /** The evaluation run ID */
-  evalRunId: import_zod27.z.string(),
+  evalRunId: import_zod26.z.string(),
   /** The scenario ID being executed */
-  scenarioId: import_zod27.z.string(),
+  scenarioId: import_zod26.z.string(),
   /** The scenario name for display */
-  scenarioName: import_zod27.z.string(),
+  scenarioName: import_zod26.z.string(),
   /** The target ID (skill, agent, etc.) */
-  targetId: import_zod27.z.string(),
+  targetId: import_zod26.z.string(),
   /** The target name for display */
-  targetName: import_zod27.z.string(),
+  targetName: import_zod26.z.string(),
   /** Step number in the current scenario execution */
-  stepNumber: import_zod27.z.number(),
+  stepNumber: import_zod26.z.number(),
   /** Type of trace event */
-  type: import_zod27.z.enum(LiveTraceEventType),
+  type: import_zod26.z.enum(LiveTraceEventType),
   /** Tool name if this is a tool_use event */
-  toolName: import_zod27.z.string().optional(),
+  toolName: import_zod26.z.string().optional(),
   /** Tool arguments preview (truncated JSON) */
-  toolArgs: import_zod27.z.string().optional(),
+  toolArgs: import_zod26.z.string().optional(),
   /** Output preview (truncated text) */
-  outputPreview: import_zod27.z.string().optional(),
+  outputPreview: import_zod26.z.string().optional(),
   /** File path for file operations */
-  filePath: import_zod27.z.string().optional(),
+  filePath: import_zod26.z.string().optional(),
   /** Elapsed time in milliseconds for progress events */
-  elapsedMs: import_zod27.z.number().optional(),
+  elapsedMs: import_zod26.z.number().optional(),
   /** Thinking/reasoning text from Claude */
-  thinking: import_zod27.z.string().optional(),
+  thinking: import_zod26.z.string().optional(),
   /** Timestamp when this event occurred */
-  timestamp: import_zod27.z.string(),
+  timestamp: import_zod26.z.string(),
   /** Whether this is the final event for this scenario */
-  isComplete: import_zod27.z.boolean()
+  isComplete: import_zod26.z.boolean()
 });
 var TRACE_EVENT_PREFIX = "TRACE_EVENT:";
 function parseTraceEventLine(line) {
@@ -2020,14 +2015,14 @@ var TriggerType = /* @__PURE__ */ ((TriggerType2) => {
   TriggerType2["MANUAL"] = "MANUAL";
   return TriggerType2;
 })(TriggerType || {});
-var TriggerMetadataSchema = import_zod28.z.object({
-  version: import_zod28.z.string().optional(),
-  resourceUpdated: import_zod28.z.array(import_zod28.z.string()).optional()
+var TriggerMetadataSchema = import_zod27.z.object({
+  version: import_zod27.z.string().optional(),
+  resourceUpdated: import_zod27.z.array(import_zod27.z.string()).optional()
 });
-var TriggerSchema = import_zod28.z.object({
-  id: import_zod28.z.string(),
+var TriggerSchema = import_zod27.z.object({
+  id: import_zod27.z.string(),
   metadata: TriggerMetadataSchema.optional(),
-  type: import_zod28.z.enum(TriggerType)
+  type: import_zod27.z.enum(TriggerType)
 });
 var FailureCategory = /* @__PURE__ */ ((FailureCategory2) => {
   FailureCategory2["MISSING_FILE"] = "missing_file";
@@ -2045,28 +2040,28 @@ var FailureSeverity = /* @__PURE__ */ ((FailureSeverity2) => {
   FailureSeverity2["LOW"] = "low";
   return FailureSeverity2;
 })(FailureSeverity || {});
-var DiffLineTypeSchema = import_zod28.z.enum(["added", "removed", "unchanged"]);
-var DiffLineSchema = import_zod28.z.object({
+var DiffLineTypeSchema = import_zod27.z.enum(["added", "removed", "unchanged"]);
+var DiffLineSchema = import_zod27.z.object({
   type: DiffLineTypeSchema,
-  content: import_zod28.z.string(),
-  lineNumber: import_zod28.z.number()
-});
-var DiffContentSchema = import_zod28.z.object({
-  path: import_zod28.z.string(),
-  expected: import_zod28.z.string(),
-  actual: import_zod28.z.string(),
-  diffLines: import_zod28.z.array(DiffLineSchema),
-  renamedFrom: import_zod28.z.string().optional()
-});
-var CommandExecutionSchema = import_zod28.z.object({
-  command: import_zod28.z.string(),
-  exitCode: import_zod28.z.number(),
-  output: import_zod28.z.string().optional(),
-  duration: import_zod28.z.number()
-});
-var FileModificationSchema = import_zod28.z.object({
-  path: import_zod28.z.string(),
-  action: import_zod28.z.enum(["created", "modified", "deleted"])
+  content: import_zod27.z.string(),
+  lineNumber: import_zod27.z.number()
+});
+var DiffContentSchema = import_zod27.z.object({
+  path: import_zod27.z.string(),
+  expected: import_zod27.z.string(),
+  actual: import_zod27.z.string(),
+  diffLines: import_zod27.z.array(DiffLineSchema),
+  renamedFrom: import_zod27.z.string().optional()
+});
+var CommandExecutionSchema = import_zod27.z.object({
+  command: import_zod27.z.string(),
+  exitCode: import_zod27.z.number(),
+  output: import_zod27.z.string().optional(),
+  duration: import_zod27.z.number()
+});
+var FileModificationSchema = import_zod27.z.object({
+  path: import_zod27.z.string(),
+  action: import_zod27.z.enum(["created", "modified", "deleted"])
 });
 var TemplateFileStatus = /* @__PURE__ */ ((TemplateFileStatus2) => {
   TemplateFileStatus2["NEW"] = "new";
@@ -2074,87 +2069,87 @@ var TemplateFileStatus = /* @__PURE__ */ ((TemplateFileStatus2) => {
   TemplateFileStatus2["UNCHANGED"] = "unchanged";
   return TemplateFileStatus2;
 })(TemplateFileStatus || {});
-var TemplateFileSchema = import_zod28.z.object({
+var TemplateFileSchema = import_zod27.z.object({
   /** Relative path within the template */
-  path: import_zod28.z.string(),
+  path: import_zod27.z.string(),
   /** Full file content after execution */
-  content: import_zod28.z.string(),
+  content: import_zod27.z.string(),
   /** File status (new, modified, unchanged) */
-  status: import_zod28.z.enum(["new", "modified", "unchanged"])
-});
-var ApiCallSchema = import_zod28.z.object({
-  endpoint: import_zod28.z.string(),
-  tokensUsed: import_zod28.z.number(),
-  duration: import_zod28.z.number()
-});
-var ExecutionTraceSchema = import_zod28.z.object({
-  commands: import_zod28.z.array(CommandExecutionSchema),
-  filesModified: import_zod28.z.array(FileModificationSchema),
-  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),
-  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(),
+  status: import_zod27.z.enum(["new", "modified", "unchanged"])
+});
+var ApiCallSchema = import_zod27.z.object({
+  endpoint: import_zod27.z.string(),
+  tokensUsed: import_zod27.z.number(),
+  duration: import_zod27.z.number()
+});
+var ExecutionTraceSchema = import_zod27.z.object({
+  commands: import_zod27.z.array(CommandExecutionSchema),
+  filesModified: import_zod27.z.array(FileModificationSchema),
+  apiCalls: import_zod27.z.array(ApiCallSchema),
+  totalDuration: import_zod27.z.number()
+});
+var FailureAnalysisSchema = import_zod27.z.object({
+  category: import_zod27.z.enum(FailureCategory),
+  severity: import_zod27.z.enum(FailureSeverity),
+  summary: import_zod27.z.string(),
+  details: import_zod27.z.string(),
+  rootCause: import_zod27.z.string(),
+  suggestedFix: import_zod27.z.string(),
+  relatedAssertions: import_zod27.z.array(import_zod27.z.string()),
+  codeSnippet: import_zod27.z.string().optional(),
+  similarIssues: import_zod27.z.array(import_zod27.z.string()).optional(),
+  patternId: import_zod27.z.string().optional(),
   // Extended fields for detailed debugging
   diff: DiffContentSchema.optional(),
   executionTrace: ExecutionTraceSchema.optional()
 });
 var EvalRunSchema = TenantEntitySchema.extend({
   /** Agent ID for this run */
-  agentId: import_zod28.z.string().optional(),
+  agentId: import_zod27.z.string().optional(),
   /** Preset ID that originated this run (optional) */
-  presetId: import_zod28.z.string().optional(),
+  presetId: import_zod27.z.string().optional(),
   /** Skill IDs for this run */
-  skillIds: import_zod28.z.array(import_zod28.z.string()).optional(),
+  skillIds: import_zod27.z.array(import_zod27.z.string()).optional(),
   /** Map of skillId to skillVersionId for this run */
-  skillVersions: import_zod28.z.record(import_zod28.z.string(), import_zod28.z.string()).optional(),
+  skillVersions: import_zod27.z.record(import_zod27.z.string(), import_zod27.z.string()).optional(),
   /** Scenario IDs to run (always present — resolved server-side from tags when needed) */
-  scenarioIds: import_zod28.z.array(import_zod28.z.string()),
+  scenarioIds: import_zod27.z.array(import_zod27.z.string()),
   /** Current status */
   status: EvalStatusSchema,
   /** Progress percentage (0-100) */
-  progress: import_zod28.z.number(),
+  progress: import_zod27.z.number(),
   /** Results for each scenario/target combination (lazy to break eval-result ↔ eval-run cycle) */
-  results: import_zod28.z.array(import_zod28.z.lazy(() => EvalRunResultSchema)),
+  results: import_zod27.z.array(import_zod27.z.lazy(() => EvalRunResultSchema)),
   /** Aggregated metrics across all results */
   aggregateMetrics: EvalMetricsSchema,
   /** Failure analyses */
-  failureAnalyses: import_zod28.z.array(FailureAnalysisSchema).optional(),
+  failureAnalyses: import_zod27.z.array(FailureAnalysisSchema).optional(),
   /** Aggregated LLM trace summary */
   llmTraceSummary: LLMTraceSummarySchema.optional(),
   /** What triggered this run */
   trigger: TriggerSchema.optional(),
   /** When the run started (set when evaluation is triggered) */
-  startedAt: import_zod28.z.string().optional(),
+  startedAt: import_zod27.z.string().optional(),
   /** When the run completed */
-  completedAt: import_zod28.z.string().optional(),
+  completedAt: import_zod27.z.string().optional(),
   /** Live trace events captured during execution (for playback on results page) */
-  liveTraceEvents: import_zod28.z.array(LiveTraceEventSchema).optional(),
+  liveTraceEvents: import_zod27.z.array(LiveTraceEventSchema).optional(),
   /** Remote job ID for tracking execution in Dev Machines */
-  jobId: import_zod28.z.string().optional(),
+  jobId: import_zod27.z.string().optional(),
   /** Remote job status from the Dev Machine API (PENDING, RUNNING, COMPLETED, FAILED, CANCELLED) */
-  jobStatus: import_zod28.z.string().optional(),
+  jobStatus: import_zod27.z.string().optional(),
   /** Remote job error message if the job failed */
-  jobError: import_zod28.z.string().optional(),
+  jobError: import_zod27.z.string().optional(),
   /** Timestamp of the last job status check */
-  jobStatusCheckedAt: import_zod28.z.string().optional(),
+  jobStatusCheckedAt: import_zod27.z.string().optional(),
   /** MCP server IDs to enable for this run (optional) */
-  mcpIds: import_zod28.z.array(import_zod28.z.string()).optional(),
+  mcpIds: import_zod27.z.array(import_zod27.z.string()).optional(),
   /** Sub-agent IDs to enable for this run (optional) */
-  subAgentIds: import_zod28.z.array(import_zod28.z.string()).optional(),
+  subAgentIds: import_zod27.z.array(import_zod27.z.string()).optional(),
   /** Rule IDs to enable for this run (optional) */
-  ruleIds: import_zod28.z.array(import_zod28.z.string()).optional(),
+  ruleIds: import_zod27.z.array(import_zod27.z.string()).optional(),
   /** Tags used to select scenarios for this run (for traceability) */
-  tags: import_zod28.z.array(import_zod28.z.string()).optional()
+  tags: import_zod27.z.array(import_zod27.z.string()).optional()
 });
 var CreateEvalRunInputSchema = EvalRunSchema.omit({
   id: true,
@@ -2169,60 +2164,60 @@ var CreateEvalRunInputSchema = EvalRunSchema.omit({
   scenarioIds: true
 }).extend({
   /** Optional on input — backend resolves from tags when not provided */
-  scenarioIds: import_zod28.z.array(import_zod28.z.string()).optional()
+  scenarioIds: import_zod27.z.array(import_zod27.z.string()).optional()
 }).refine(
   (data) => data.scenarioIds && data.scenarioIds.length > 0 || data.tags && data.tags.length > 0,
   { message: "Either scenarioIds or tags must be provided" }
 );
-var EvaluationProgressSchema = import_zod28.z.object({
-  runId: import_zod28.z.string(),
-  targetId: import_zod28.z.string(),
-  totalScenarios: import_zod28.z.number(),
-  completedScenarios: import_zod28.z.number(),
-  scenarioProgress: import_zod28.z.array(
-    import_zod28.z.object({
-      scenarioId: import_zod28.z.string(),
-      currentStep: import_zod28.z.string(),
-      error: import_zod28.z.string().optional()
+var EvaluationProgressSchema = import_zod27.z.object({
+  runId: import_zod27.z.string(),
+  targetId: import_zod27.z.string(),
+  totalScenarios: import_zod27.z.number(),
+  completedScenarios: import_zod27.z.number(),
+  scenarioProgress: import_zod27.z.array(
+    import_zod27.z.object({
+      scenarioId: import_zod27.z.string(),
+      currentStep: import_zod27.z.string(),
+      error: import_zod27.z.string().optional()
     })
   ),
-  createdAt: import_zod28.z.number()
-});
-var EvaluationLogSchema = import_zod28.z.object({
-  runId: import_zod28.z.string(),
-  scenarioId: import_zod28.z.string(),
-  log: import_zod28.z.object({
-    level: import_zod28.z.enum(["info", "error", "debug"]),
-    message: import_zod28.z.string().optional(),
-    args: import_zod28.z.array(import_zod28.z.any()).optional(),
-    error: import_zod28.z.string().optional()
+  createdAt: import_zod27.z.number()
+});
+var EvaluationLogSchema = import_zod27.z.object({
+  runId: import_zod27.z.string(),
+  scenarioId: import_zod27.z.string(),
+  log: import_zod27.z.object({
+    level: import_zod27.z.enum(["info", "error", "debug"]),
+    message: import_zod27.z.string().optional(),
+    args: import_zod27.z.array(import_zod27.z.any()).optional(),
+    error: import_zod27.z.string().optional()
   })
 });
 var LLM_TIMEOUT = 12e4;
 
 // src/evaluation/conversation.ts
-var import_zod29 = require("zod");
-var TextBlockSchema = import_zod29.z.object({
-  type: import_zod29.z.literal("text"),
-  text: import_zod29.z.string()
-});
-var ThinkingBlockSchema = import_zod29.z.object({
-  type: import_zod29.z.literal("thinking"),
-  thinking: import_zod29.z.string()
-});
-var ToolUseBlockSchema = import_zod29.z.object({
-  type: import_zod29.z.literal("tool_use"),
-  toolName: import_zod29.z.string(),
-  toolId: import_zod29.z.string(),
-  input: import_zod29.z.unknown()
-});
-var ToolResultBlockSchema = import_zod29.z.object({
-  type: import_zod29.z.literal("tool_result"),
-  toolUseId: import_zod29.z.string(),
-  content: import_zod29.z.string(),
-  isError: import_zod29.z.boolean().optional()
-});
-var ConversationBlockSchema = import_zod29.z.discriminatedUnion("type", [
+var import_zod28 = require("zod");
+var TextBlockSchema = import_zod28.z.object({
+  type: import_zod28.z.literal("text"),
+  text: import_zod28.z.string()
+});
+var ThinkingBlockSchema = import_zod28.z.object({
+  type: import_zod28.z.literal("thinking"),
+  thinking: import_zod28.z.string()
+});
+var ToolUseBlockSchema = import_zod28.z.object({
+  type: import_zod28.z.literal("tool_use"),
+  toolName: import_zod28.z.string(),
+  toolId: import_zod28.z.string(),
+  input: import_zod28.z.unknown()
+});
+var ToolResultBlockSchema = import_zod28.z.object({
+  type: import_zod28.z.literal("tool_result"),
+  toolUseId: import_zod28.z.string(),
+  content: import_zod28.z.string(),
+  isError: import_zod28.z.boolean().optional()
+});
+var ConversationBlockSchema = import_zod28.z.discriminatedUnion("type", [
   TextBlockSchema,
   ThinkingBlockSchema,
   ToolUseBlockSchema,
@@ -2233,18 +2228,18 @@ var ConversationMessageRoles = [
   "user",
   "system"
 ];
-var ConversationMessageSchema = import_zod29.z.object({
-  role: import_zod29.z.enum(ConversationMessageRoles),
-  content: import_zod29.z.array(ConversationBlockSchema),
-  timestamp: import_zod29.z.string()
+var ConversationMessageSchema = import_zod28.z.object({
+  role: import_zod28.z.enum(ConversationMessageRoles),
+  content: import_zod28.z.array(ConversationBlockSchema),
+  timestamp: import_zod28.z.string()
 });
-var ScenarioConversationSchema = import_zod29.z.object({
-  id: import_zod29.z.string(),
-  projectId: import_zod29.z.string(),
-  evalRunId: import_zod29.z.string(),
-  resultId: import_zod29.z.string(),
-  messages: import_zod29.z.array(ConversationMessageSchema),
-  createdAt: import_zod29.z.string()
+var ScenarioConversationSchema = import_zod28.z.object({
+  id: import_zod28.z.string(),
+  projectId: import_zod28.z.string(),
+  evalRunId: import_zod28.z.string(),
+  resultId: import_zod28.z.string(),
+  messages: import_zod28.z.array(ConversationMessageSchema),
+  createdAt: import_zod28.z.string()
 });
 
 // src/evaluation/eval-result.ts
@@ -2255,100 +2250,100 @@ var AssertionResultStatus = /* @__PURE__ */ ((AssertionResultStatus2) => {
   AssertionResultStatus2["ERROR"] = "error";
   return AssertionResultStatus2;
 })(AssertionResultStatus || {});
-var AssertionResultSchema = import_zod30.z.object({
-  id: import_zod30.z.string(),
-  assertionId: import_zod30.z.string(),
-  assertionType: import_zod30.z.string(),
-  assertionName: import_zod30.z.string(),
-  status: import_zod30.z.enum(AssertionResultStatus),
-  message: import_zod30.z.string().optional(),
-  expected: import_zod30.z.string().optional(),
-  actual: import_zod30.z.string().optional(),
-  duration: import_zod30.z.number().optional(),
-  details: import_zod30.z.record(import_zod30.z.string(), import_zod30.z.unknown()).optional(),
-  llmTraceSteps: import_zod30.z.array(LLMTraceStepSchema).optional()
-});
-var EvalRunResultSchema = import_zod30.z.object({
-  id: import_zod30.z.string(),
-  targetId: import_zod30.z.string(),
-  targetName: import_zod30.z.string().optional(),
+var AssertionResultSchema = import_zod29.z.object({
+  id: import_zod29.z.string(),
+  assertionId: import_zod29.z.string(),
+  assertionType: import_zod29.z.string(),
+  assertionName: import_zod29.z.string(),
+  status: import_zod29.z.enum(AssertionResultStatus),
+  message: import_zod29.z.string().optional(),
+  expected: import_zod29.z.string().optional(),
+  actual: import_zod29.z.string().optional(),
+  duration: import_zod29.z.number().optional(),
+  details: import_zod29.z.record(import_zod29.z.string(), import_zod29.z.unknown()).optional(),
+  llmTraceSteps: import_zod29.z.array(LLMTraceStepSchema).optional()
+});
+var EvalRunResultSchema = import_zod29.z.object({
+  id: import_zod29.z.string(),
+  targetId: import_zod29.z.string(),
+  targetName: import_zod29.z.string().optional(),
   /** SkillVersion ID used for this evaluation (for version tracking) */
-  skillVersionId: import_zod30.z.string().optional(),
+  skillVersionId: import_zod29.z.string().optional(),
   /** SkillVersion semver string (e.g., "1.0.0", "1.2.3") for display */
-  skillVersion: import_zod30.z.string().optional(),
-  scenarioId: import_zod30.z.string(),
-  scenarioName: import_zod30.z.string(),
+  skillVersion: import_zod29.z.string().optional(),
+  scenarioId: import_zod29.z.string(),
+  scenarioName: import_zod29.z.string(),
   modelConfig: ModelConfigSchema.optional(),
-  assertionResults: import_zod30.z.array(AssertionResultSchema),
+  assertionResults: import_zod29.z.array(AssertionResultSchema),
   metrics: EvalMetricsSchema.optional(),
-  passed: import_zod30.z.number(),
-  failed: import_zod30.z.number(),
-  passRate: import_zod30.z.number(),
-  duration: import_zod30.z.number(),
-  outputText: import_zod30.z.string().optional(),
-  files: import_zod30.z.array(ExpectedFileSchema).optional(),
-  fileDiffs: import_zod30.z.array(DiffContentSchema).optional(),
+  passed: import_zod29.z.number(),
+  failed: import_zod29.z.number(),
+  passRate: import_zod29.z.number(),
+  duration: import_zod29.z.number(),
+  outputText: import_zod29.z.string().optional(),
+  files: import_zod29.z.array(ExpectedFileSchema).optional(),
+  fileDiffs: import_zod29.z.array(DiffContentSchema).optional(),
   /** Full template files after execution with status indicators */
-  templateFiles: import_zod30.z.array(TemplateFileSchema).optional(),
-  startedAt: import_zod30.z.string().optional(),
-  completedAt: import_zod30.z.string().optional(),
+  templateFiles: import_zod29.z.array(TemplateFileSchema).optional(),
+  startedAt: import_zod29.z.string().optional(),
+  completedAt: import_zod29.z.string().optional(),
   llmTrace: LLMTraceSchema.optional(),
   /** Full conversation messages (only present in transit; stripped before DB storage) */
-  conversation: import_zod30.z.array(ConversationMessageSchema).optional()
-});
-var PromptResultSchema = import_zod30.z.object({
-  text: import_zod30.z.string(),
-  files: import_zod30.z.array(import_zod30.z.unknown()).optional(),
-  finishReason: import_zod30.z.string().optional(),
-  reasoning: import_zod30.z.string().optional(),
-  reasoningDetails: import_zod30.z.unknown().optional(),
-  toolCalls: import_zod30.z.array(import_zod30.z.unknown()).optional(),
-  toolResults: import_zod30.z.array(import_zod30.z.unknown()).optional(),
-  warnings: import_zod30.z.array(import_zod30.z.unknown()).optional(),
-  sources: import_zod30.z.array(import_zod30.z.unknown()).optional(),
-  steps: import_zod30.z.array(import_zod30.z.unknown()),
-  generationTimeMs: import_zod30.z.number(),
-  prompt: import_zod30.z.string(),
-  systemPrompt: import_zod30.z.string(),
-  usage: import_zod30.z.object({
-    totalTokens: import_zod30.z.number().optional(),
-    totalMicrocentsSpent: import_zod30.z.number().optional()
+  conversation: import_zod29.z.array(ConversationMessageSchema).optional()
+});
+var PromptResultSchema = import_zod29.z.object({
+  text: import_zod29.z.string(),
+  files: import_zod29.z.array(import_zod29.z.unknown()).optional(),
+  finishReason: import_zod29.z.string().optional(),
+  reasoning: import_zod29.z.string().optional(),
+  reasoningDetails: import_zod29.z.unknown().optional(),
+  toolCalls: import_zod29.z.array(import_zod29.z.unknown()).optional(),
+  toolResults: import_zod29.z.array(import_zod29.z.unknown()).optional(),
+  warnings: import_zod29.z.array(import_zod29.z.unknown()).optional(),
+  sources: import_zod29.z.array(import_zod29.z.unknown()).optional(),
+  steps: import_zod29.z.array(import_zod29.z.unknown()),
+  generationTimeMs: import_zod29.z.number(),
+  prompt: import_zod29.z.string(),
+  systemPrompt: import_zod29.z.string(),
+  usage: import_zod29.z.object({
+    totalTokens: import_zod29.z.number().optional(),
+    totalMicrocentsSpent: import_zod29.z.number().optional()
   })
 });
-var EvaluationResultSchema = import_zod30.z.object({
-  id: import_zod30.z.string(),
-  runId: import_zod30.z.string(),
-  timestamp: import_zod30.z.number(),
+var EvaluationResultSchema = import_zod29.z.object({
+  id: import_zod29.z.string(),
+  runId: import_zod29.z.string(),
+  timestamp: import_zod29.z.number(),
   promptResult: PromptResultSchema,
-  testResults: import_zod30.z.array(import_zod30.z.unknown()),
-  tags: import_zod30.z.array(import_zod30.z.string()).optional(),
-  feedback: import_zod30.z.string().optional(),
-  score: import_zod30.z.number(),
-  suiteId: import_zod30.z.string().optional()
-});
-var LeanEvaluationResultSchema = import_zod30.z.object({
-  id: import_zod30.z.string(),
-  runId: import_zod30.z.string(),
-  timestamp: import_zod30.z.number(),
-  tags: import_zod30.z.array(import_zod30.z.string()).optional(),
-  scenarioId: import_zod30.z.string(),
-  scenarioVersion: import_zod30.z.number().optional(),
-  targetId: import_zod30.z.string(),
-  targetVersion: import_zod30.z.number().optional(),
-  suiteId: import_zod30.z.string().optional(),
-  score: import_zod30.z.number(),
-  time: import_zod30.z.number().optional(),
-  microcentsSpent: import_zod30.z.number().optional()
+  testResults: import_zod29.z.array(import_zod29.z.unknown()),
+  tags: import_zod29.z.array(import_zod29.z.string()).optional(),
+  feedback: import_zod29.z.string().optional(),
+  score: import_zod29.z.number(),
+  suiteId: import_zod29.z.string().optional()
+});
+var LeanEvaluationResultSchema = import_zod29.z.object({
+  id: import_zod29.z.string(),
+  runId: import_zod29.z.string(),
+  timestamp: import_zod29.z.number(),
+  tags: import_zod29.z.array(import_zod29.z.string()).optional(),
+  scenarioId: import_zod29.z.string(),
+  scenarioVersion: import_zod29.z.number().optional(),
+  targetId: import_zod29.z.string(),
+  targetVersion: import_zod29.z.number().optional(),
+  suiteId: import_zod29.z.string().optional(),
+  score: import_zod29.z.number(),
+  time: import_zod29.z.number().optional(),
+  microcentsSpent: import_zod29.z.number().optional()
 });
 
 // src/project/project.ts
-var import_zod31 = require("zod");
+var import_zod30 = require("zod");
 var ProjectSchema = BaseEntitySchema.extend({
-  appId: import_zod31.z.string().optional().describe("The ID of the app in Dev Center"),
-  appSecret: import_zod31.z.string().optional().describe("The secret of the app in Dev Center"),
-  useWixAuth: import_zod31.z.boolean().optional().describe("Enable Wix CLI/MCP auth for evaluations"),
-  useBase44Auth: import_zod31.z.boolean().optional().describe("Enable Base44 auth for evaluations"),
-  scenarioTags: import_zod31.z.array(import_zod31.z.string()).optional().describe("Project-level tag vocabulary for scenarios")
+  appId: import_zod30.z.string().optional().describe("The ID of the app in Dev Center"),
+  appSecret: import_zod30.z.string().optional().describe("The secret of the app in Dev Center"),
+  useWixAuth: import_zod30.z.boolean().optional().describe("Enable Wix CLI/MCP auth for evaluations"),
+  useBase44Auth: import_zod30.z.boolean().optional().describe("Enable Base44 auth for evaluations"),
+  scenarioTags: import_zod30.z.array(import_zod30.z.string()).optional().describe("Project-level tag vocabulary for scenarios")
 });
 var CreateProjectInputSchema = ProjectSchema.omit({
   id: true,
@@ -2401,7 +2396,7 @@ var SYSTEM_ASSERTIONS = {
   [SYSTEM_ASSERTION_IDS.TOOL_CALLED_WITH_PARAM]: {
     id: SYSTEM_ASSERTION_IDS.TOOL_CALLED_WITH_PARAM,
     name: "Tool Called With Param",
-    description: "Check that a tool was called with expected parameters",
+    description: "Check that a tool was called with expected parameters (tool name is substring matched)",
     type: "tool_called_with_param",
     parameters: [
       {
@@ -2415,6 +2410,14 @@ var SYSTEM_ASSERTIONS = {
         label: "Expected Parameters (JSON, substring match)",
         type: "string",
         required: true
+      },
+      {
+        name: "requireSuccess",
+        label: "Require Successful Call",
+        type: "boolean",
+        required: false,
+        defaultValue: false,
+        advanced: true
       }
     ]
   },
@@ -2547,6 +2550,9 @@ function getSystemAssertion(id) {
   BuildCheckTestSchema,
   BuildPassedAssertionSchema,
   BuildPassedConfigSchema,
+  BulkImportResultItemSchema,
+  BulkImportResultSchema,
+  BulkImportSkillsInputSchema,
   ClaudeModel,
   ClaudeModelSchema,
   CommandExecutionSchema,