@wix/evalforge-types 0.45.0 → 0.47.0

package/build/index.mjs

@@ -929,17 +929,17 @@ var WebhookIdentityType;
 })(WebhookIdentityType || (WebhookIdentityType = {}));
 
 // src/common/models.ts
-var AVAILABLE_MODEL_IDS = Object.values(
+var AVAILABLE_CLAUDE_MODEL_IDS = Object.values(
   ClaudeModel
 ).filter(
   (v) => typeof v === "string" && v !== ClaudeModel.UNKNOWN_CLAUDE_MODEL
 );
 var PREFERRED_JUDGE_MODEL = "CLAUDE_4_5_HAIKU_1_0";
-var DEFAULT_JUDGE_MODEL = AVAILABLE_MODEL_IDS.includes(
+var DEFAULT_JUDGE_MODEL = AVAILABLE_CLAUDE_MODEL_IDS.includes(
   PREFERRED_JUDGE_MODEL
-) ? PREFERRED_JUDGE_MODEL : AVAILABLE_MODEL_IDS[0];
+) ? PREFERRED_JUDGE_MODEL : AVAILABLE_CLAUDE_MODEL_IDS[0];
 var ClaudeModelSchema = z4.enum(
-  AVAILABLE_MODEL_IDS
+  AVAILABLE_CLAUDE_MODEL_IDS
 );
 var AVAILABLE_OPENAI_MODEL_IDS = Object.values(
   Model
@@ -950,7 +950,7 @@ var OpenAIModelSchema = z4.enum(
   AVAILABLE_OPENAI_MODEL_IDS
 );
 var ALL_AVAILABLE_MODEL_IDS = [
-  ...AVAILABLE_MODEL_IDS,
+  ...AVAILABLE_CLAUDE_MODEL_IDS,
   ...AVAILABLE_OPENAI_MODEL_IDS
 ];
 var AnyModelSchema = z4.enum(
@@ -1697,7 +1697,7 @@ var LLMTraceSchema = z26.object({
 });
 
 // src/evaluation/eval-result.ts
-import { z as z29 } from "zod";
+import { z as z30 } from "zod";
 
 // src/evaluation/eval-run.ts
 import { z as z28 } from "zod";
@@ -1944,6 +1944,53 @@ var EvaluationLogSchema = z28.object({
 });
 var LLM_TIMEOUT = 12e4;
 
+// src/evaluation/conversation.ts
+import { z as z29 } from "zod";
+var TextBlockSchema = z29.object({
+  type: z29.literal("text"),
+  text: z29.string()
+});
+var ThinkingBlockSchema = z29.object({
+  type: z29.literal("thinking"),
+  thinking: z29.string()
+});
+var ToolUseBlockSchema = z29.object({
+  type: z29.literal("tool_use"),
+  toolName: z29.string(),
+  toolId: z29.string(),
+  input: z29.unknown()
+});
+var ToolResultBlockSchema = z29.object({
+  type: z29.literal("tool_result"),
+  toolUseId: z29.string(),
+  content: z29.string(),
+  isError: z29.boolean().optional()
+});
+var ConversationBlockSchema = z29.discriminatedUnion("type", [
+  TextBlockSchema,
+  ThinkingBlockSchema,
+  ToolUseBlockSchema,
+  ToolResultBlockSchema
+]);
+var ConversationMessageRoles = [
+  "assistant",
+  "user",
+  "system"
+];
+var ConversationMessageSchema = z29.object({
+  role: z29.enum(ConversationMessageRoles),
+  content: z29.array(ConversationBlockSchema),
+  timestamp: z29.string()
+});
+var ScenarioConversationSchema = z29.object({
+  id: z29.string(),
+  projectId: z29.string(),
+  evalRunId: z29.string(),
+  resultId: z29.string(),
+  messages: z29.array(ConversationMessageSchema),
+  createdAt: z29.string()
+});
+
 // src/evaluation/eval-result.ts
 var AssertionResultStatus = /* @__PURE__ */ ((AssertionResultStatus2) => {
   AssertionResultStatus2["PASSED"] = "passed";
@@ -1952,97 +1999,99 @@ var AssertionResultStatus = /* @__PURE__ */ ((AssertionResultStatus2) => {
   AssertionResultStatus2["ERROR"] = "error";
   return AssertionResultStatus2;
 })(AssertionResultStatus || {});
-var AssertionResultSchema = z29.object({
-  id: z29.string(),
-  assertionId: z29.string(),
-  assertionType: z29.string(),
-  assertionName: z29.string(),
-  status: z29.enum(AssertionResultStatus),
-  message: z29.string().optional(),
-  expected: z29.string().optional(),
-  actual: z29.string().optional(),
-  duration: z29.number().optional(),
-  details: z29.record(z29.string(), z29.unknown()).optional(),
-  llmTraceSteps: z29.array(LLMTraceStepSchema).optional()
-});
-var EvalRunResultSchema = z29.object({
-  id: z29.string(),
-  targetId: z29.string(),
-  targetName: z29.string().optional(),
+var AssertionResultSchema = z30.object({
+  id: z30.string(),
+  assertionId: z30.string(),
+  assertionType: z30.string(),
+  assertionName: z30.string(),
+  status: z30.enum(AssertionResultStatus),
+  message: z30.string().optional(),
+  expected: z30.string().optional(),
+  actual: z30.string().optional(),
+  duration: z30.number().optional(),
+  details: z30.record(z30.string(), z30.unknown()).optional(),
+  llmTraceSteps: z30.array(LLMTraceStepSchema).optional()
+});
+var EvalRunResultSchema = z30.object({
+  id: z30.string(),
+  targetId: z30.string(),
+  targetName: z30.string().optional(),
   /** SkillVersion ID used for this evaluation (for version tracking) */
-  skillVersionId: z29.string().optional(),
+  skillVersionId: z30.string().optional(),
   /** SkillVersion semver string (e.g., "1.0.0", "1.2.3") for display */
-  skillVersion: z29.string().optional(),
-  scenarioId: z29.string(),
-  scenarioName: z29.string(),
+  skillVersion: z30.string().optional(),
+  scenarioId: z30.string(),
+  scenarioName: z30.string(),
   modelConfig: ModelConfigSchema.optional(),
-  assertionResults: z29.array(AssertionResultSchema),
+  assertionResults: z30.array(AssertionResultSchema),
   metrics: EvalMetricsSchema.optional(),
-  passed: z29.number(),
-  failed: z29.number(),
-  passRate: z29.number(),
-  duration: z29.number(),
-  outputText: z29.string().optional(),
-  files: z29.array(ExpectedFileSchema).optional(),
-  fileDiffs: z29.array(DiffContentSchema).optional(),
+  passed: z30.number(),
+  failed: z30.number(),
+  passRate: z30.number(),
+  duration: z30.number(),
+  outputText: z30.string().optional(),
+  files: z30.array(ExpectedFileSchema).optional(),
+  fileDiffs: z30.array(DiffContentSchema).optional(),
   /** Full template files after execution with status indicators */
-  templateFiles: z29.array(TemplateFileSchema).optional(),
-  startedAt: z29.string().optional(),
-  completedAt: z29.string().optional(),
-  llmTrace: LLMTraceSchema.optional()
-});
-var PromptResultSchema = z29.object({
-  text: z29.string(),
-  files: z29.array(z29.unknown()).optional(),
-  finishReason: z29.string().optional(),
-  reasoning: z29.string().optional(),
-  reasoningDetails: z29.unknown().optional(),
-  toolCalls: z29.array(z29.unknown()).optional(),
-  toolResults: z29.array(z29.unknown()).optional(),
-  warnings: z29.array(z29.unknown()).optional(),
-  sources: z29.array(z29.unknown()).optional(),
-  steps: z29.array(z29.unknown()),
-  generationTimeMs: z29.number(),
-  prompt: z29.string(),
-  systemPrompt: z29.string(),
-  usage: z29.object({
-    totalTokens: z29.number().optional(),
-    totalMicrocentsSpent: z29.number().optional()
+  templateFiles: z30.array(TemplateFileSchema).optional(),
+  startedAt: z30.string().optional(),
+  completedAt: z30.string().optional(),
+  llmTrace: LLMTraceSchema.optional(),
+  /** Full conversation messages (only present in transit; stripped before DB storage) */
+  conversation: z30.array(ConversationMessageSchema).optional()
+});
+var PromptResultSchema = z30.object({
+  text: z30.string(),
+  files: z30.array(z30.unknown()).optional(),
+  finishReason: z30.string().optional(),
+  reasoning: z30.string().optional(),
+  reasoningDetails: z30.unknown().optional(),
+  toolCalls: z30.array(z30.unknown()).optional(),
+  toolResults: z30.array(z30.unknown()).optional(),
+  warnings: z30.array(z30.unknown()).optional(),
+  sources: z30.array(z30.unknown()).optional(),
+  steps: z30.array(z30.unknown()),
+  generationTimeMs: z30.number(),
+  prompt: z30.string(),
+  systemPrompt: z30.string(),
+  usage: z30.object({
+    totalTokens: z30.number().optional(),
+    totalMicrocentsSpent: z30.number().optional()
   })
 });
-var EvaluationResultSchema = z29.object({
-  id: z29.string(),
-  runId: z29.string(),
-  timestamp: z29.number(),
+var EvaluationResultSchema = z30.object({
+  id: z30.string(),
+  runId: z30.string(),
+  timestamp: z30.number(),
   promptResult: PromptResultSchema,
-  testResults: z29.array(z29.unknown()),
-  tags: z29.array(z29.string()).optional(),
-  feedback: z29.string().optional(),
-  score: z29.number(),
-  suiteId: z29.string().optional()
-});
-var LeanEvaluationResultSchema = z29.object({
-  id: z29.string(),
-  runId: z29.string(),
-  timestamp: z29.number(),
-  tags: z29.array(z29.string()).optional(),
-  scenarioId: z29.string(),
-  scenarioVersion: z29.number().optional(),
-  targetId: z29.string(),
-  targetVersion: z29.number().optional(),
-  suiteId: z29.string().optional(),
-  score: z29.number(),
-  time: z29.number().optional(),
-  microcentsSpent: z29.number().optional()
+  testResults: z30.array(z30.unknown()),
+  tags: z30.array(z30.string()).optional(),
+  feedback: z30.string().optional(),
+  score: z30.number(),
+  suiteId: z30.string().optional()
+});
+var LeanEvaluationResultSchema = z30.object({
+  id: z30.string(),
+  runId: z30.string(),
+  timestamp: z30.number(),
+  tags: z30.array(z30.string()).optional(),
+  scenarioId: z30.string(),
+  scenarioVersion: z30.number().optional(),
+  targetId: z30.string(),
+  targetVersion: z30.number().optional(),
+  suiteId: z30.string().optional(),
+  score: z30.number(),
+  time: z30.number().optional(),
+  microcentsSpent: z30.number().optional()
 });
 
 // src/project/project.ts
-import { z as z30 } from "zod";
+import { z as z31 } from "zod";
 var ProjectSchema = BaseEntitySchema.extend({
-  appId: z30.string().optional().describe("The ID of the app in Dev Center"),
-  appSecret: z30.string().optional().describe("The secret of the app in Dev Center"),
-  useWixAuth: z30.boolean().optional().describe("Enable Wix CLI/MCP auth for evaluations"),
-  useBase44Auth: z30.boolean().optional().describe("Enable Base44 auth for evaluations")
+  appId: z31.string().optional().describe("The ID of the app in Dev Center"),
+  appSecret: z31.string().optional().describe("The secret of the app in Dev Center"),
+  useWixAuth: z31.boolean().optional().describe("Enable Wix CLI/MCP auth for evaluations"),
+  useBase44Auth: z31.boolean().optional().describe("Enable Base44 auth for evaluations")
 });
 var CreateProjectInputSchema = ProjectSchema.omit({
   id: true,
@@ -2216,7 +2265,7 @@ function getSystemAssertion(id) {
 export {
   AGENT_TYPE_LABELS,
   ALL_AVAILABLE_MODEL_IDS,
-  AVAILABLE_MODEL_IDS,
+  AVAILABLE_CLAUDE_MODEL_IDS,
   AVAILABLE_OPENAI_MODEL_IDS,
   AVAILABLE_RUN_COMMANDS,
   AVAILABLE_TOOL_NAMES,
@@ -2244,6 +2293,9 @@ export {
   ClaudeModelSchema,
   CommandExecutionSchema,
   CommandExecutionTestSchema,
+  ConversationBlockSchema,
+  ConversationMessageRoles,
+  ConversationMessageSchema,
   CostAssertionSchema,
   CostConfigSchema,
   CreateAgentInputSchema,
@@ -2316,6 +2368,7 @@ export {
   SYSTEM_ASSERTIONS,
   SYSTEM_ASSERTION_IDS,
   ScenarioAssertionLinkSchema,
+  ScenarioConversationSchema,
   SiteConfigTestSchema,
   SkillFileSchema,
   SkillMetadataSchema,
@@ -2340,12 +2393,16 @@ export {
   TestSuiteSchema,
   TestType,
   TestTypeSchema,
+  TextBlockSchema,
+  ThinkingBlockSchema,
   TimeAssertionSchema,
   TimeConfigSchema,
   TokenUsageSchema,
   ToolCalledWithParamAssertionSchema,
   ToolCalledWithParamConfigSchema,
+  ToolResultBlockSchema,
   ToolTestSchema,
+  ToolUseBlockSchema,
   TriggerMetadataSchema,
   TriggerSchema,
   TriggerType,