@wix/evalforge-types 0.44.0 → 0.46.0

package/build/index.js

@@ -50,6 +50,9 @@ __export(index_exports, {
   ClaudeModelSchema: () => ClaudeModelSchema,
   CommandExecutionSchema: () => CommandExecutionSchema,
   CommandExecutionTestSchema: () => CommandExecutionTestSchema,
+  ConversationBlockSchema: () => ConversationBlockSchema,
+  ConversationMessageRoles: () => ConversationMessageRoles,
+  ConversationMessageSchema: () => ConversationMessageSchema,
   CostAssertionSchema: () => CostAssertionSchema,
   CostConfigSchema: () => CostConfigSchema,
   CreateAgentInputSchema: () => CreateAgentInputSchema,
@@ -122,6 +125,7 @@ __export(index_exports, {
   SYSTEM_ASSERTIONS: () => SYSTEM_ASSERTIONS,
   SYSTEM_ASSERTION_IDS: () => SYSTEM_ASSERTION_IDS,
   ScenarioAssertionLinkSchema: () => ScenarioAssertionLinkSchema,
+  ScenarioConversationSchema: () => ScenarioConversationSchema,
   SiteConfigTestSchema: () => SiteConfigTestSchema,
   SkillFileSchema: () => SkillFileSchema,
   SkillMetadataSchema: () => SkillMetadataSchema,
@@ -146,12 +150,16 @@ __export(index_exports, {
   TestSuiteSchema: () => TestSuiteSchema,
   TestType: () => TestType,
   TestTypeSchema: () => TestTypeSchema,
+  TextBlockSchema: () => TextBlockSchema,
+  ThinkingBlockSchema: () => ThinkingBlockSchema,
   TimeAssertionSchema: () => TimeAssertionSchema,
   TimeConfigSchema: () => TimeConfigSchema,
   TokenUsageSchema: () => TokenUsageSchema,
   ToolCalledWithParamAssertionSchema: () => ToolCalledWithParamAssertionSchema,
   ToolCalledWithParamConfigSchema: () => ToolCalledWithParamConfigSchema,
+  ToolResultBlockSchema: () => ToolResultBlockSchema,
   ToolTestSchema: () => ToolTestSchema,
+  ToolUseBlockSchema: () => ToolUseBlockSchema,
   TriggerMetadataSchema: () => TriggerMetadataSchema,
   TriggerSchema: () => TriggerSchema,
   TriggerType: () => TriggerType,
@@ -1879,7 +1887,7 @@ var LLMTraceSchema = import_zod26.z.object({
 });
 
 // src/evaluation/eval-result.ts
-var import_zod29 = require("zod");
+var import_zod30 = require("zod");
 
 // src/evaluation/eval-run.ts
 var import_zod28 = require("zod");
@@ -2126,6 +2134,53 @@ var EvaluationLogSchema = import_zod28.z.object({
 });
 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", [
+  TextBlockSchema,
+  ThinkingBlockSchema,
+  ToolUseBlockSchema,
+  ToolResultBlockSchema
+]);
+var ConversationMessageRoles = [
+  "assistant",
+  "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 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()
+});
+
 // src/evaluation/eval-result.ts
 var AssertionResultStatus = /* @__PURE__ */ ((AssertionResultStatus2) => {
   AssertionResultStatus2["PASSED"] = "passed";
@@ -2134,95 +2189,99 @@ var AssertionResultStatus = /* @__PURE__ */ ((AssertionResultStatus2) => {
   AssertionResultStatus2["ERROR"] = "error";
   return AssertionResultStatus2;
 })(AssertionResultStatus || {});
-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(),
+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(),
   /** SkillVersion ID used for this evaluation (for version tracking) */
-  skillVersionId: import_zod29.z.string().optional(),
+  skillVersionId: import_zod30.z.string().optional(),
   /** SkillVersion semver string (e.g., "1.0.0", "1.2.3") for display */
-  skillVersion: import_zod29.z.string().optional(),
-  scenarioId: import_zod29.z.string(),
-  scenarioName: import_zod29.z.string(),
+  skillVersion: import_zod30.z.string().optional(),
+  scenarioId: import_zod30.z.string(),
+  scenarioName: import_zod30.z.string(),
   modelConfig: ModelConfigSchema.optional(),
-  assertionResults: import_zod29.z.array(AssertionResultSchema),
+  assertionResults: import_zod30.z.array(AssertionResultSchema),
   metrics: EvalMetricsSchema.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(),
+  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(),
   /** Full template files after execution with status indicators */
-  templateFiles: import_zod29.z.array(TemplateFileSchema).optional(),
-  startedAt: import_zod29.z.string().optional(),
-  completedAt: import_zod29.z.string().optional(),
-  llmTrace: LLMTraceSchema.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()
+  templateFiles: import_zod30.z.array(TemplateFileSchema).optional(),
+  startedAt: import_zod30.z.string().optional(),
+  completedAt: import_zod30.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()
   })
 });
-var EvaluationResultSchema = import_zod29.z.object({
-  id: import_zod29.z.string(),
-  runId: import_zod29.z.string(),
-  timestamp: import_zod29.z.number(),
+var EvaluationResultSchema = import_zod30.z.object({
+  id: import_zod30.z.string(),
+  runId: import_zod30.z.string(),
+  timestamp: import_zod30.z.number(),
   promptResult: PromptResultSchema,
-  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()
+  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()
 });
 
 // src/project/project.ts
-var import_zod30 = require("zod");
+var import_zod31 = require("zod");
 var ProjectSchema = BaseEntitySchema.extend({
-  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")
+  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")
 });
 var CreateProjectInputSchema = ProjectSchema.omit({
   id: true,
@@ -2425,6 +2484,9 @@ function getSystemAssertion(id) {
   ClaudeModelSchema,
   CommandExecutionSchema,
   CommandExecutionTestSchema,
+  ConversationBlockSchema,
+  ConversationMessageRoles,
+  ConversationMessageSchema,
   CostAssertionSchema,
   CostConfigSchema,
   CreateAgentInputSchema,
@@ -2497,6 +2559,7 @@ function getSystemAssertion(id) {
   SYSTEM_ASSERTIONS,
   SYSTEM_ASSERTION_IDS,
   ScenarioAssertionLinkSchema,
+  ScenarioConversationSchema,
   SiteConfigTestSchema,
   SkillFileSchema,
   SkillMetadataSchema,
@@ -2521,12 +2584,16 @@ function getSystemAssertion(id) {
   TestSuiteSchema,
   TestType,
   TestTypeSchema,
+  TextBlockSchema,
+  ThinkingBlockSchema,
   TimeAssertionSchema,
   TimeConfigSchema,
   TokenUsageSchema,
   ToolCalledWithParamAssertionSchema,
   ToolCalledWithParamConfigSchema,
+  ToolResultBlockSchema,
   ToolTestSchema,
+  ToolUseBlockSchema,
   TriggerMetadataSchema,
   TriggerSchema,
   TriggerType,