@wix/evalforge-evaluator 0.186.0 → 0.188.0

package/build/types/ambassador-converters.d.ts

@@ -30,7 +30,8 @@ import type { TestScenario as WireTestScenario } from '@wix/ambassador-evalforge
 import type { Template as WireTemplate } from '@wix/ambassador-evalforge-v1-project/types';
 import type { Preset as WirePreset } from '@wix/ambassador-evalforge-v1-preset/types';
 import type { Capability as WireCapability, CapabilityVersion as WireCapabilityVersion } from '@wix/ambassador-evalforge-v1-capability/types';
-import type { Agent, Capability, CapabilityVersion, EvalRun, EvalRunResult, LiveTraceEvent, Preset, Template, TestScenario } from '@wix/evalforge-types';
+import type { ProvisionedSite as WireProvisionedSite, SiteBootstrapResult as WireSiteBootstrapResult } from '@wix/ambassador-evalforge-v1-site-provisioning/types';
+import type { Agent, Capability, CapabilityVersion, EvalRun, EvalRunResult, LiveTraceEvent, Preset, ProvisionedSite, SiteBootstrapResult, Template, TestScenario } from '@wix/evalforge-types';
 export declare function evalRunResultToProto(result: EvalRunResult): WireEvalRunResult;
 export declare function liveTraceEventToProto(event: LiveTraceEvent): WireLiveTraceEvent;
 export declare function agentFromProto(wire: WireAgent): Agent;
@@ -66,3 +67,9 @@ export declare function capabilityFromProto(wire: WireCapability): Capability;
  * by the parent capability's `capabilityType`.
  */
 export declare function capabilityVersionFromProto(wire: WireCapabilityVersion, projectId: string): CapabilityVersion;
+export declare function provisionedSiteFromProto(proto: WireProvisionedSite): ProvisionedSite;
+/**
+ * Map the wire bootstrap result onto the domain shape. Returns `undefined`
+ * when no bootstrap ran (no result on the response).
+ */
+export declare function siteBootstrapResultFromProto(proto: WireSiteBootstrapResult | undefined | null): SiteBootstrapResult | undefined;

package/build/types/api-client.d.ts

@@ -11,7 +11,7 @@
  * (status/completedAt/jobError/jobStatus) the evaluator writes.
  */
 import { type IHttpClient } from '@wix/http-client';
-import type { EvalRun, EvalRunResult, LiveTraceEvent, TestScenario, Template, Agent, Preset, CapabilityWithLatestVersion, CapabilityVersion } from '@wix/evalforge-types';
+import type { EvalRun, EvalRunResult, LiveTraceEvent, TestScenario, Template, Agent, Preset, CapabilityWithLatestVersion, CapabilityVersion, ProvisionedSite } from '@wix/evalforge-types';
 export interface ApiClient {
     getEvalRun(projectId: string, id: string): Promise<EvalRun>;
     getScenario(projectId: string, id: string): Promise<TestScenario>;
@@ -24,6 +24,8 @@ export interface ApiClient {
     clearResults(projectId: string, evalRunId: string): Promise<void>;
     pushTraceEvent(projectId: string, evalRunId: string, event: LiveTraceEvent): Promise<void>;
     updateEvalRun(projectId: string, evalRunId: string, update: Partial<EvalRun>): Promise<void>;
+    provisionScenarioSite(projectId: string, evalRunId: string, scenarioId: string): Promise<ProvisionedSite>;
+    deleteProvisionedSite(projectId: string, siteId: string): Promise<void>;
 }
 /**
  * Options for creating an API client.

package/build/types/run-scenario/environment.d.ts

@@ -1,5 +1,8 @@
 import type { Template } from '@wix/evalforge-types';
 import type { EvaluatorConfig } from '../config.js';
+export interface PrepareWorkingDirectoryOptions {
+    template?: Template;
+}
 /**
  * Prepare the working directory for a scenario.
  * If a template with source is provided, fetch files from GitHub.
@@ -9,7 +12,8 @@ import type { EvaluatorConfig } from '../config.js';
  * @param evalRunId - The evaluation run ID
  * @param targetId - The target ID (prompt agent, skill, or agent)
  * @param scenarioId - The scenario ID (for per-scenario isolation when no template)
- * @param template - Optional pre-fetched template entity
+ * @param onProgress - Required callback invoked with human-readable setup progress messages
+ * @param options - Optional settings: template (pre-fetched template entity)
  * @returns The working directory path
  */
-export declare function prepareWorkingDirectory(config: EvaluatorConfig, evalRunId: string, targetId: string, scenarioId: string, template?: Template): Promise<string>;
+export declare function prepareWorkingDirectory(config: EvaluatorConfig, evalRunId: string, targetId: string, scenarioId: string, onProgress: (message: string) => void, options?: PrepareWorkingDirectoryOptions): Promise<string>;

package/build/types/run-scenario/index.d.ts

@@ -2,6 +2,7 @@ import type { EvalRunResult, LiveTraceEvent, TestScenario, Template } from '@wix
 import type { EvaluatorConfig } from '../config.js';
 import type { Assertion } from '@wix/evalforge-types';
 import type { EvaluationData } from '../fetch-evaluation-data.js';
+import type { ApiClient } from '../api-client.js';
 export type { PartialEvalRunResult } from './types.js';
 export type { ScenarioItem, EvaluationData } from '../fetch-evaluation-data.js';
 /**
@@ -17,4 +18,4 @@ export type { ScenarioItem, EvaluationData } from '../fetch-evaluation-data.js';
  *   (remote job execution, where stdout is not accessible)
  * @returns Complete evaluation result
  */
-export declare function runScenario(config: EvaluatorConfig, evalRunId: string, scenario: TestScenario, evalData: EvaluationData, template?: Template, resolvedAssertions?: Assertion[], pushEvent?: (event: LiveTraceEvent) => void): Promise<EvalRunResult>;
+export declare function runScenario(config: EvaluatorConfig, evalRunId: string, scenario: TestScenario, evalData: EvaluationData, template?: Template, resolvedAssertions?: Assertion[], pushEvent?: (event: LiveTraceEvent) => void, apiClient?: ApiClient, projectId?: string): Promise<EvalRunResult>;

package/build/types/run-scenario/install-dependencies.d.ts

@@ -1,4 +1,8 @@
 import { execFileSync } from 'child_process';
+export interface InstallDependenciesOptions {
+    exec?: typeof execFileSync;
+    cacheBase?: string;
+}
 /**
  * Install dependencies in workDir if a package.json is present.
  * Detects the package manager from the lock file and runs the appropriate install command.
@@ -7,5 +11,9 @@ import { execFileSync } from 'child_process';
  * When cacheBase is provided, node_modules is cached keyed by the lock file hash.
  * Subsequent calls with the same lock file content skip the install entirely —
  * node_modules is restored via APFS clone on macOS (near-instant) or recursive copy on Linux.
+ *
+ * @param workDir - The directory in which to install dependencies
+ * @param onProgress - Callback invoked with a human-readable message when an install runs
+ * @param options - Optional settings: exec override and cacheBase directory
  */
-export declare function installDependencies(workDir: string, exec?: typeof execFileSync, cacheBase?: string): Promise<void>;
+export declare function installDependencies(workDir: string, onProgress: (message: string) => void, options?: InstallDependenciesOptions): Promise<void>;

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@wix/evalforge-evaluator",
-  "version": "0.186.0",
+  "version": "0.188.0",
   "description": "EvalForge Evaluator",
   "bin": "./build/index.js",
   "files": [
@@ -22,9 +22,9 @@
     "@ai-sdk/openai": "^3.0.39",
     "@anthropic-ai/claude-agent-sdk": "^0.2.63",
     "@anthropic-ai/claude-code": "^2.1.63",
-    "@wix/eval-assertions": "0.70.0",
-    "@wix/evalforge-github-client": "0.70.0",
-    "@wix/evalforge-types": "0.95.0",
+    "@wix/eval-assertions": "0.71.0",
+    "@wix/evalforge-github-client": "0.71.0",
+    "@wix/evalforge-types": "0.96.0",
     "ai": "^6.0.107",
     "diff": "^7.0.0",
     "tar": "^7.5.3",
@@ -40,7 +40,8 @@
     "@wix/ambassador-evalforge-v1-eval-run": "^1.0.8",
     "@wix/ambassador-evalforge-v1-preset": "^1.0.5",
     "@wix/ambassador-evalforge-v1-project": "^1.0.6",
-    "@wix/ambassador-evalforge-v1-test-scenario": "^1.0.5",
+    "@wix/ambassador-evalforge-v1-site-provisioning": "^1.0.3",
+    "@wix/ambassador-evalforge-v1-test-scenario": "^1.0.6",
     "@wix/http-client": "^2.85.0",
     "@wix/http-client-testkit": "^1.764.0",
     "dotenv": "^17.2.3",
@@ -71,5 +72,5 @@
       "artifactId": "evalforge-evaluator"
     }
   },
-  "falconPackageHash": "5f4f2c58e85af8542618199134badc85b38cc4d069ded5f71c47c7b7"
+  "falconPackageHash": "b8b148914d1fa0beef82451a376a65ce77a0c5aa77199a1f55f12f63"
 }