feat(import): add evaluator and online eval config import subcommands

src/cli/commands/import/__tests__/import-online-eval.test.ts

@@ -0,0 +1,368 @@
+/**
+ * Import Online Eval Config Unit Tests
+ *
+ * Covers:
+ * - extractAgentName: service name parsing
+ * - toOnlineEvalConfigSpec conversion: happy path, missing sampling, enableOnCreate
+ * - Template logical ID lookup for online eval configs
+ * - Phase 2 import resource list construction for online eval configs
+ */
+import type { GetOnlineEvalConfigResult } from '../../../aws/agentcore-control';
+import { extractAgentName, toOnlineEvalConfigSpec } from '../import-online-eval';
+import { buildImportTemplate, findLogicalIdByProperty, findLogicalIdsByType } from '../template-utils';
+import type { CfnTemplate } from '../template-utils';
+import type { ResourceToImport } from '../types';
+import { describe, expect, it } from 'vitest';
+
+// ============================================================================
+// extractAgentName Tests
+// ============================================================================
+
+describe('extractAgentName', () => {
+ it('extracts agent name from service name with .DEFAULT suffix', () => {
+ expect(extractAgentName(['my_agent.DEFAULT'])).toBe('my_agent');
+ });
+
+ it('extracts agent name with project prefix pattern', () => {
+ expect(extractAgentName(['testproject_my_agent.DEFAULT'])).toBe('testproject_my_agent');
+ });
+
+ it('returns full string when no dot suffix', () => {
+ expect(extractAgentName(['my_agent'])).toBe('my_agent');
+ });
+
+ it('returns undefined for empty array', () => {
+ expect(extractAgentName([])).toBeUndefined();
+ });
+
+ it('uses first service name when multiple provided', () => {
+ expect(extractAgentName(['agent_one.DEFAULT', 'agent_two.DEFAULT'])).toBe('agent_one');
+ });
+
+ it('handles service name with multiple dots', () => {
+ expect(extractAgentName(['my.agent.DEFAULT'])).toBe('my.agent');
+ });
+});
+
+// ============================================================================
+// toOnlineEvalConfigSpec Conversion Tests
+// ============================================================================
+
+describe('toOnlineEvalConfigSpec', () => {
+ it('maps online eval config with all fields', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-123',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-123',
+ configName: 'QualityMonitor',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ description: 'Monitor agent quality',
+ samplingPercentage: 50,
+ serviceNames: ['my_agent.DEFAULT'],
+ evaluatorIds: ['eval-456'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'QualityMonitor', 'my_agent', ['my_evaluator']);
+
+ expect(result.name).toBe('QualityMonitor');
+ expect(result.agent).toBe('my_agent');
+ expect(result.evaluators).toEqual(['my_evaluator']);
+ expect(result.samplingRate).toBe(50);
+ expect(result.description).toBe('Monitor agent quality');
+ expect(result.enableOnCreate).toBe(true);
+ });
+
+ it('omits enableOnCreate when execution status is DISABLED', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-456',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-456',
+ configName: 'DisabledConfig',
+ status: 'ACTIVE',
+ executionStatus: 'DISABLED',
+ samplingPercentage: 10,
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'DisabledConfig', 'agent', ['eval_one']);
+
+ expect(result.enableOnCreate).toBeUndefined();
+ });
+
+ it('omits description when not present', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-789',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-789',
+ configName: 'NoDesc',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ samplingPercentage: 25,
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'NoDesc', 'agent', ['eval_one']);
+
+ expect(result.description).toBeUndefined();
+ });
+
+ it('throws when sampling percentage is missing', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-no-sampling',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-no-sampling',
+ configName: 'NoSampling',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1'],
+ };
+
+ expect(() => toOnlineEvalConfigSpec(detail, 'NoSampling', 'agent', ['eval_one'])).toThrow(
+ 'has no sampling configuration'
+ );
+ });
+
+ it('supports multiple evaluator references', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-multi',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-multi',
+ configName: 'MultiEval',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ samplingPercentage: 75,
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1', 'eval-2'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'MultiEval', 'agent', [
+ 'local_eval',
+ 'arn:aws:bedrock-agentcore:us-west-2:123456789012:evaluator/eval-2',
+ ]);
+
+ expect(result.evaluators).toHaveLength(2);
+ expect(result.evaluators[0]).toBe('local_eval');
+ expect(result.evaluators[1]).toMatch(/^arn:/);
+ });
+});
+
+// ============================================================================
+// Template Logical ID Lookup Tests for Online Eval Configs
+// ============================================================================
+
+describe('Template Logical ID Lookup for Online Eval Configs', () => {
+ const synthTemplate: CfnTemplate = {
+ Resources: {
+ MyOnlineEvalConfig: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'QualityMonitor',
+ },
+ },
+ PrefixedOnlineEvalConfig: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'TestProject_PrefixedConfig',
+ },
+ },
+ MyAgentRuntime: {
+ Type: 'AWS::BedrockAgentCore::Runtime',
+ Properties: {
+ AgentRuntimeName: 'TestProject_my_agent',
+ },
+ },
+ MyIAMRole: {
+ Type: 'AWS::IAM::Role',
+ Properties: {
+ RoleName: 'MyRole',
+ },
+ },
+ },
+ };
+
+ it('finds online eval config logical ID by OnlineEvaluationConfigName property', () => {
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'QualityMonitor'
+ );
+ expect(logicalId).toBe('MyOnlineEvalConfig');
+ });
+
+ it('finds prefixed online eval config by full name', () => {
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'TestProject_PrefixedConfig'
+ );
+ expect(logicalId).toBe('PrefixedOnlineEvalConfig');
+ });
+
+ it('finds all online eval config logical IDs by type', () => {
+ const logicalIds = findLogicalIdsByType(synthTemplate, 'AWS::BedrockAgentCore::OnlineEvaluationConfig');
+ expect(logicalIds).toHaveLength(2);
+ expect(logicalIds).toContain('MyOnlineEvalConfig');
+ expect(logicalIds).toContain('PrefixedOnlineEvalConfig');
+ });
+
+ it('returns undefined for non-existent config name', () => {
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'nonexistent_config'
+ );
+ expect(logicalId).toBeUndefined();
+ });
+
+ it('falls back to single online eval config logical ID when name does not match', () => {
+ const singleConfigTemplate: CfnTemplate = {
+ Resources: {
+ OnlyConfig: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'some_config',
+ },
+ },
+ },
+ };
+
+ let logicalId = findLogicalIdByProperty(
+ singleConfigTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'different_name'
+ );
+
+ // Primary lookup fails
+ expect(logicalId).toBeUndefined();
+
+ // Fallback: if there's only one config resource, use it
+ if (!logicalId) {
+ const configLogicalIds = findLogicalIdsByType(
+ singleConfigTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig'
+ );
+ if (configLogicalIds.length === 1) {
+ logicalId = configLogicalIds[0];
+ }
+ }
+ expect(logicalId).toBe('OnlyConfig');
+ });
+});
+
+// ============================================================================
+// Phase 2 Resource Import List Construction for Online Eval Configs
+// ============================================================================
+
+describe('Phase 2: ResourceToImport List Construction for Online Eval Configs', () => {
+ const synthTemplate: CfnTemplate = {
+ Resources: {
+ OnlineEvalLogicalId: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'QualityMonitor',
+ },
+ },
+ IAMRoleLogicalId: {
+ Type: 'AWS::IAM::Role',
+ Properties: {},
+ },
+ },
+ };
+
+ it('builds ResourceToImport list for online eval config', () => {
+ const configName = 'QualityMonitor';
+ const configId = 'oec-123';
+
+ const resourcesToImport: ResourceToImport[] = [];
+
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ configName
+ );
+
+ if (logicalId) {
+ resourcesToImport.push({
+ resourceType: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ logicalResourceId: logicalId,
+ resourceIdentifier: { OnlineEvaluationConfigId: configId },
+ });
+ }
+
+ expect(resourcesToImport).toHaveLength(1);
+ expect(resourcesToImport[0]!.resourceType).toBe('AWS::BedrockAgentCore::OnlineEvaluationConfig');
+ expect(resourcesToImport[0]!.logicalResourceId).toBe('OnlineEvalLogicalId');
+ expect(resourcesToImport[0]!.resourceIdentifier).toEqual({ OnlineEvaluationConfigId: 'oec-123' });
+ });
+
+ it('returns empty list when online eval config not found in template', () => {
+ const emptyTemplate: CfnTemplate = {
+ Resources: {
+ IAMRoleLogicalId: {
+ Type: 'AWS::IAM::Role',
+ Properties: {},
+ },
+ },
+ };
+
+ const logicalId = findLogicalIdByProperty(
+ emptyTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'QualityMonitor'
+ );
+
+ expect(logicalId).toBeUndefined();
+ });
+});
+
+// ============================================================================
+// buildImportTemplate Tests for Online Eval Config Resources
+// ============================================================================
+
+describe('buildImportTemplate with Online Eval Config', () => {
+ it('adds online eval config resource to deployed template with Retain deletion policy', () => {
+ const deployedTemplate: CfnTemplate = {
+ Resources: {
+ ExistingIAMRole: {
+ Type: 'AWS::IAM::Role',
+ Properties: { RoleName: 'ExistingRole' },
+ },
+ },
+ };
+
+ const synthTemplate: CfnTemplate = {
+ Resources: {
+ ExistingIAMRole: {
+ Type: 'AWS::IAM::Role',
+ Properties: { RoleName: 'ExistingRole' },
+ },
+ OnlineEvalLogicalId: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'QualityMonitor',
+ },
+ DependsOn: 'ExistingIAMRole',
+ },
+ },
+ };
+
+ const importTemplate = buildImportTemplate(deployedTemplate, synthTemplate, ['OnlineEvalLogicalId']);
+
+ // Verify online eval config resource was added
+ expect(importTemplate.Resources.OnlineEvalLogicalId).toBeDefined();
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.Type).toBe('AWS::BedrockAgentCore::OnlineEvaluationConfig');
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.DeletionPolicy).toBe('Retain');
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.UpdateReplacePolicy).toBe('Retain');
+
+ // DependsOn should be removed for import
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.DependsOn).toBeUndefined();
+
+ // Original resource should still be there
+ expect(importTemplate.Resources.ExistingIAMRole).toBeDefined();
+ });
+});

src/cli/commands/import/command.ts

@@ -1,14 +1,13 @@
 import { handleImport } from './actions';
+import { ANSI } from './constants';
+import { registerImportEvaluator } from './import-evaluator';
 import { registerImportMemory } from './import-memory';
+import { registerImportOnlineEval } from './import-online-eval';
 import { registerImportRuntime } from './import-runtime';
 import type { Command } from '@commander-js/extra-typings';
 import * as fs from 'node:fs';
 
-const green = '\x1b[32m';
-const yellow = '\x1b[33m';
-const cyan = '\x1b[36m';
-const dim = '\x1b[2m';
-const reset = '\x1b[0m';
+const { green, yellow, cyan, dim, reset } = ANSI;
 
 export const registerImport = (program: Command) => {
  const importCmd = program
@@ -151,4 +150,6 @@ export const registerImport = (program: Command) => {
  // Register subcommands for importing individual resource types from AWS
  registerImportRuntime(importCmd);
  registerImportMemory(importCmd);
+ registerImportEvaluator(importCmd);
+ registerImportOnlineEval(importCmd);
 };

src/cli/commands/import/__tests__/import-online-eval.test.ts

@@ -0,0 +1,368 @@
+/**
+ * Import Online Eval Config Unit Tests
+ *
+ * Covers:
+ * - extractAgentName: service name parsing
+ * - toOnlineEvalConfigSpec conversion: happy path, missing sampling, enableOnCreate
+ * - Template logical ID lookup for online eval configs
+ * - Phase 2 import resource list construction for online eval configs
+ */
+import type { GetOnlineEvalConfigResult } from '../../../aws/agentcore-control';
+import { extractAgentName, toOnlineEvalConfigSpec } from '../import-online-eval';
+import { buildImportTemplate, findLogicalIdByProperty, findLogicalIdsByType } from '../template-utils';
+import type { CfnTemplate } from '../template-utils';
+import type { ResourceToImport } from '../types';
+import { describe, expect, it } from 'vitest';
+
+// ============================================================================
+// extractAgentName Tests
+// ============================================================================
+
+describe('extractAgentName', () => {
+ it('extracts agent name from service name with .DEFAULT suffix', () => {
+ expect(extractAgentName(['my_agent.DEFAULT'])).toBe('my_agent');
+ });
+
+ it('extracts agent name with project prefix pattern', () => {
+ expect(extractAgentName(['testproject_my_agent.DEFAULT'])).toBe('testproject_my_agent');
+ });
+
+ it('returns full string when no dot suffix', () => {
+ expect(extractAgentName(['my_agent'])).toBe('my_agent');
+ });
+
+ it('returns undefined for empty array', () => {
+ expect(extractAgentName([])).toBeUndefined();
+ });
+
+ it('uses first service name when multiple provided', () => {
+ expect(extractAgentName(['agent_one.DEFAULT', 'agent_two.DEFAULT'])).toBe('agent_one');
+ });
+
+ it('handles service name with multiple dots', () => {
+ expect(extractAgentName(['my.agent.DEFAULT'])).toBe('my.agent');
+ });
+});
+
+// ============================================================================
+// toOnlineEvalConfigSpec Conversion Tests
+// ============================================================================
+
+describe('toOnlineEvalConfigSpec', () => {
+ it('maps online eval config with all fields', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-123',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-123',
+ configName: 'QualityMonitor',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ description: 'Monitor agent quality',
+ samplingPercentage: 50,
+ serviceNames: ['my_agent.DEFAULT'],
+ evaluatorIds: ['eval-456'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'QualityMonitor', 'my_agent', ['my_evaluator']);
+
+ expect(result.name).toBe('QualityMonitor');
+ expect(result.agent).toBe('my_agent');
+ expect(result.evaluators).toEqual(['my_evaluator']);
+ expect(result.samplingRate).toBe(50);
+ expect(result.description).toBe('Monitor agent quality');
+ expect(result.enableOnCreate).toBe(true);
+ });
+
+ it('omits enableOnCreate when execution status is DISABLED', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-456',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-456',
+ configName: 'DisabledConfig',
+ status: 'ACTIVE',
+ executionStatus: 'DISABLED',
+ samplingPercentage: 10,
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'DisabledConfig', 'agent', ['eval_one']);
+
+ expect(result.enableOnCreate).toBeUndefined();
+ });
+
+ it('omits description when not present', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-789',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-789',
+ configName: 'NoDesc',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ samplingPercentage: 25,
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'NoDesc', 'agent', ['eval_one']);
+
+ expect(result.description).toBeUndefined();
+ });
+
+ it('throws when sampling percentage is missing', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-no-sampling',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-no-sampling',
+ configName: 'NoSampling',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1'],
+ };
+
+ expect(() => toOnlineEvalConfigSpec(detail, 'NoSampling', 'agent', ['eval_one'])).toThrow(
+ 'has no sampling configuration'
+ );
+ });
+
+ it('supports multiple evaluator references', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-multi',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-multi',
+ configName: 'MultiEval',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ samplingPercentage: 75,
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1', 'eval-2'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'MultiEval', 'agent', [
+ 'local_eval',
+ 'arn:aws:bedrock-agentcore:us-west-2:123456789012:evaluator/eval-2',
+ ]);
+
+ expect(result.evaluators).toHaveLength(2);
+ expect(result.evaluators[0]).toBe('local_eval');
+ expect(result.evaluators[1]).toMatch(/^arn:/);
+ });
+});
+
+// ============================================================================
+// Template Logical ID Lookup Tests for Online Eval Configs
+// ============================================================================
+
+describe('Template Logical ID Lookup for Online Eval Configs', () => {
+ const synthTemplate: CfnTemplate = {
+ Resources: {
+ MyOnlineEvalConfig: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'QualityMonitor',
+ },
+ },
+ PrefixedOnlineEvalConfig: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'TestProject_PrefixedConfig',
+ },
+ },
+ MyAgentRuntime: {
+ Type: 'AWS::BedrockAgentCore::Runtime',
+ Properties: {
+ AgentRuntimeName: 'TestProject_my_agent',
+ },
+ },
+ MyIAMRole: {
+ Type: 'AWS::IAM::Role',
+ Properties: {
+ RoleName: 'MyRole',
+ },
+ },
+ },
+ };
+
+ it('finds online eval config logical ID by OnlineEvaluationConfigName property', () => {
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'QualityMonitor'
+ );
+ expect(logicalId).toBe('MyOnlineEvalConfig');
+ });
+
+ it('finds prefixed online eval config by full name', () => {
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'TestProject_PrefixedConfig'
+ );
+ expect(logicalId).toBe('PrefixedOnlineEvalConfig');
+ });
+
+ it('finds all online eval config logical IDs by type', () => {
+ const logicalIds = findLogicalIdsByType(synthTemplate, 'AWS::BedrockAgentCore::OnlineEvaluationConfig');
+ expect(logicalIds).toHaveLength(2);
+ expect(logicalIds).toContain('MyOnlineEvalConfig');
+ expect(logicalIds).toContain('PrefixedOnlineEvalConfig');
+ });
+
+ it('returns undefined for non-existent config name', () => {
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'nonexistent_config'
+ );
+ expect(logicalId).toBeUndefined();
+ });
+
+ it('falls back to single online eval config logical ID when name does not match', () => {
+ const singleConfigTemplate: CfnTemplate = {
+ Resources: {
+ OnlyConfig: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'some_config',
+ },
+ },
+ },
+ };
+
+ let logicalId = findLogicalIdByProperty(
+ singleConfigTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'different_name'
+ );
+
+ // Primary lookup fails
+ expect(logicalId).toBeUndefined();
+
+ // Fallback: if there's only one config resource, use it
+ if (!logicalId) {
+ const configLogicalIds = findLogicalIdsByType(
+ singleConfigTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig'
+ );
+ if (configLogicalIds.length === 1) {
+ logicalId = configLogicalIds[0];
+ }
+ }
+ expect(logicalId).toBe('OnlyConfig');
+ });
+});
+
+// ============================================================================
+// Phase 2 Resource Import List Construction for Online Eval Configs
+// ============================================================================
+
+describe('Phase 2: ResourceToImport List Construction for Online Eval Configs', () => {
+ const synthTemplate: CfnTemplate = {
+ Resources: {
+ OnlineEvalLogicalId: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'QualityMonitor',
+ },
+ },
+ IAMRoleLogicalId: {
+ Type: 'AWS::IAM::Role',
+ Properties: {},
+ },
+ },
+ };
+
+ it('builds ResourceToImport list for online eval config', () => {
+ const configName = 'QualityMonitor';
+ const configId = 'oec-123';
+
+ const resourcesToImport: ResourceToImport[] = [];
+
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ configName
+ );
+
+ if (logicalId) {
+ resourcesToImport.push({
+ resourceType: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ logicalResourceId: logicalId,
+ resourceIdentifier: { OnlineEvaluationConfigId: configId },
+ });
+ }
+
+ expect(resourcesToImport).toHaveLength(1);
+ expect(resourcesToImport[0]!.resourceType).toBe('AWS::BedrockAgentCore::OnlineEvaluationConfig');
+ expect(resourcesToImport[0]!.logicalResourceId).toBe('OnlineEvalLogicalId');
+ expect(resourcesToImport[0]!.resourceIdentifier).toEqual({ OnlineEvaluationConfigId: 'oec-123' });
+ });
+
+ it('returns empty list when online eval config not found in template', () => {
+ const emptyTemplate: CfnTemplate = {
+ Resources: {
+ IAMRoleLogicalId: {
+ Type: 'AWS::IAM::Role',
+ Properties: {},
+ },
+ },
+ };
+
+ const logicalId = findLogicalIdByProperty(
+ emptyTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'QualityMonitor'
+ );
+
+ expect(logicalId).toBeUndefined();
+ });
+});
+
+// ============================================================================
+// buildImportTemplate Tests for Online Eval Config Resources
+// ============================================================================
+
+describe('buildImportTemplate with Online Eval Config', () => {
+ it('adds online eval config resource to deployed template with Retain deletion policy', () => {
+ const deployedTemplate: CfnTemplate = {
+ Resources: {
+ ExistingIAMRole: {
+ Type: 'AWS::IAM::Role',
+ Properties: { RoleName: 'ExistingRole' },
+ },
+ },
+ };
+
+ const synthTemplate: CfnTemplate = {
+ Resources: {
+ ExistingIAMRole: {
+ Type: 'AWS::IAM::Role',
+ Properties: { RoleName: 'ExistingRole' },
+ },
+ OnlineEvalLogicalId: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'QualityMonitor',
+ },
+ DependsOn: 'ExistingIAMRole',
+ },
+ },
+ };
+
+ const importTemplate = buildImportTemplate(deployedTemplate, synthTemplate, ['OnlineEvalLogicalId']);
+
+ // Verify online eval config resource was added
+ expect(importTemplate.Resources.OnlineEvalLogicalId).toBeDefined();
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.Type).toBe('AWS::BedrockAgentCore::OnlineEvaluationConfig');
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.DeletionPolicy).toBe('Retain');
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.UpdateReplacePolicy).toBe('Retain');
+
+ // DependsOn should be removed for import
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.DependsOn).toBeUndefined();
+
+ // Original resource should still be there
+ expect(importTemplate.Resources.ExistingIAMRole).toBeDefined();
+ });
+});

src/cli/commands/import/command.ts

@@ -1,14 +1,13 @@
 import { handleImport } from './actions';
+import { ANSI } from './constants';
+import { registerImportEvaluator } from './import-evaluator';
 import { registerImportMemory } from './import-memory';
+import { registerImportOnlineEval } from './import-online-eval';
 import { registerImportRuntime } from './import-runtime';
 import type { Command } from '@commander-js/extra-typings';
 import * as fs from 'node:fs';
 
-const green = '\x1b[32m';
-const yellow = '\x1b[33m';
-const cyan = '\x1b[36m';
-const dim = '\x1b[2m';
-const reset = '\x1b[0m';
+const { green, yellow, cyan, dim, reset } = ANSI;
 
 export const registerImport = (program: Command) => {
  const importCmd = program
@@ -151,4 +150,6 @@ export const registerImport = (program: Command) => {
  // Register subcommands for importing individual resource types from AWS
  registerImportRuntime(importCmd);
  registerImportMemory(importCmd);
+ registerImportEvaluator(importCmd);
+ registerImportOnlineEval(importCmd);
 };

src/cli/commands/import/__tests__/import-online-eval.test.ts

@@ -0,0 +1,368 @@
+/**
+ * Import Online Eval Config Unit Tests
+ *
+ * Covers:
+ * - extractAgentName: service name parsing
+ * - toOnlineEvalConfigSpec conversion: happy path, missing sampling, enableOnCreate
+ * - Template logical ID lookup for online eval configs
+ * - Phase 2 import resource list construction for online eval configs
+ */
+import type { GetOnlineEvalConfigResult } from '../../../aws/agentcore-control';
+import { extractAgentName, toOnlineEvalConfigSpec } from '../import-online-eval';
+import { buildImportTemplate, findLogicalIdByProperty, findLogicalIdsByType } from '../template-utils';
+import type { CfnTemplate } from '../template-utils';
+import type { ResourceToImport } from '../types';
+import { describe, expect, it } from 'vitest';
+
+// ============================================================================
+// extractAgentName Tests
+// ============================================================================
+
+describe('extractAgentName', () => {
+ it('extracts agent name from service name with .DEFAULT suffix', () => {
+ expect(extractAgentName(['my_agent.DEFAULT'])).toBe('my_agent');
+ });
+
+ it('extracts agent name with project prefix pattern', () => {
+ expect(extractAgentName(['testproject_my_agent.DEFAULT'])).toBe('testproject_my_agent');
+ });
+
+ it('returns full string when no dot suffix', () => {
+ expect(extractAgentName(['my_agent'])).toBe('my_agent');
+ });
+
+ it('returns undefined for empty array', () => {
+ expect(extractAgentName([])).toBeUndefined();
+ });
+
+ it('uses first service name when multiple provided', () => {
+ expect(extractAgentName(['agent_one.DEFAULT', 'agent_two.DEFAULT'])).toBe('agent_one');
+ });
+
+ it('handles service name with multiple dots', () => {
+ expect(extractAgentName(['my.agent.DEFAULT'])).toBe('my.agent');
+ });
+});
+
+// ============================================================================
+// toOnlineEvalConfigSpec Conversion Tests
+// ============================================================================
+
+describe('toOnlineEvalConfigSpec', () => {
+ it('maps online eval config with all fields', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-123',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-123',
+ configName: 'QualityMonitor',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ description: 'Monitor agent quality',
+ samplingPercentage: 50,
+ serviceNames: ['my_agent.DEFAULT'],
+ evaluatorIds: ['eval-456'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'QualityMonitor', 'my_agent', ['my_evaluator']);
+
+ expect(result.name).toBe('QualityMonitor');
+ expect(result.agent).toBe('my_agent');
+ expect(result.evaluators).toEqual(['my_evaluator']);
+ expect(result.samplingRate).toBe(50);
+ expect(result.description).toBe('Monitor agent quality');
+ expect(result.enableOnCreate).toBe(true);
+ });
+
+ it('omits enableOnCreate when execution status is DISABLED', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-456',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-456',
+ configName: 'DisabledConfig',
+ status: 'ACTIVE',
+ executionStatus: 'DISABLED',
+ samplingPercentage: 10,
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'DisabledConfig', 'agent', ['eval_one']);
+
+ expect(result.enableOnCreate).toBeUndefined();
+ });
+
+ it('omits description when not present', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-789',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-789',
+ configName: 'NoDesc',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ samplingPercentage: 25,
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'NoDesc', 'agent', ['eval_one']);
+
+ expect(result.description).toBeUndefined();
+ });
+
+ it('throws when sampling percentage is missing', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-no-sampling',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-no-sampling',
+ configName: 'NoSampling',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1'],
+ };
+
+ expect(() => toOnlineEvalConfigSpec(detail, 'NoSampling', 'agent', ['eval_one'])).toThrow(
+ 'has no sampling configuration'
+ );
+ });
+
+ it('supports multiple evaluator references', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-multi',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-multi',
+ configName: 'MultiEval',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ samplingPercentage: 75,
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1', 'eval-2'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'MultiEval', 'agent', [
+ 'local_eval',
+ 'arn:aws:bedrock-agentcore:us-west-2:123456789012:evaluator/eval-2',
+ ]);
+
+ expect(result.evaluators).toHaveLength(2);
+ expect(result.evaluators[0]).toBe('local_eval');
+ expect(result.evaluators[1]).toMatch(/^arn:/);
+ });
+});
+
+// ============================================================================
+// Template Logical ID Lookup Tests for Online Eval Configs
+// ============================================================================
+
+describe('Template Logical ID Lookup for Online Eval Configs', () => {
+ const synthTemplate: CfnTemplate = {
+ Resources: {
+ MyOnlineEvalConfig: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'QualityMonitor',
+ },
+ },
+ PrefixedOnlineEvalConfig: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'TestProject_PrefixedConfig',
+ },
+ },
+ MyAgentRuntime: {
+ Type: 'AWS::BedrockAgentCore::Runtime',
+ Properties: {
+ AgentRuntimeName: 'TestProject_my_agent',
+ },
+ },
+ MyIAMRole: {
+ Type: 'AWS::IAM::Role',
+ Properties: {
+ RoleName: 'MyRole',
+ },
+ },
+ },
+ };
+
+ it('finds online eval config logical ID by OnlineEvaluationConfigName property', () => {
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'QualityMonitor'
+ );
+ expect(logicalId).toBe('MyOnlineEvalConfig');
+ });
+
+ it('finds prefixed online eval config by full name', () => {
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'TestProject_PrefixedConfig'
+ );
+ expect(logicalId).toBe('PrefixedOnlineEvalConfig');
+ });
+
+ it('finds all online eval config logical IDs by type', () => {
+ const logicalIds = findLogicalIdsByType(synthTemplate, 'AWS::BedrockAgentCore::OnlineEvaluationConfig');
+ expect(logicalIds).toHaveLength(2);
+ expect(logicalIds).toContain('MyOnlineEvalConfig');
+ expect(logicalIds).toContain('PrefixedOnlineEvalConfig');
+ });
+
+ it('returns undefined for non-existent config name', () => {
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'nonexistent_config'
+ );
+ expect(logicalId).toBeUndefined();
+ });
+
+ it('falls back to single online eval config logical ID when name does not match', () => {
+ const singleConfigTemplate: CfnTemplate = {
+ Resources: {
+ OnlyConfig: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'some_config',
+ },
+ },
+ },
+ };
+
+ let logicalId = findLogicalIdByProperty(
+ singleConfigTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'different_name'
+ );
+
+ // Primary lookup fails
+ expect(logicalId).toBeUndefined();
+
+ // Fallback: if there's only one config resource, use it
+ if (!logicalId) {
+ const configLogicalIds = findLogicalIdsByType(
+ singleConfigTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig'
+ );
+ if (configLogicalIds.length === 1) {
+ logicalId = configLogicalIds[0];
+ }
+ }
+ expect(logicalId).toBe('OnlyConfig');
+ });
+});
+
+// ============================================================================
+// Phase 2 Resource Import List Construction for Online Eval Configs
+// ============================================================================
+
+describe('Phase 2: ResourceToImport List Construction for Online Eval Configs', () => {
+ const synthTemplate: CfnTemplate = {
+ Resources: {
+ OnlineEvalLogicalId: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'QualityMonitor',
+ },
+ },
+ IAMRoleLogicalId: {
+ Type: 'AWS::IAM::Role',
+ Properties: {},
+ },
+ },
+ };
+
+ it('builds ResourceToImport list for online eval config', () => {
+ const configName = 'QualityMonitor';
+ const configId = 'oec-123';
+
+ const resourcesToImport: ResourceToImport[] = [];
+
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ configName
+ );
+
+ if (logicalId) {
+ resourcesToImport.push({
+ resourceType: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ logicalResourceId: logicalId,
+ resourceIdentifier: { OnlineEvaluationConfigId: configId },
+ });
+ }
+
+ expect(resourcesToImport).toHaveLength(1);
+ expect(resourcesToImport[0]!.resourceType).toBe('AWS::BedrockAgentCore::OnlineEvaluationConfig');
+ expect(resourcesToImport[0]!.logicalResourceId).toBe('OnlineEvalLogicalId');
+ expect(resourcesToImport[0]!.resourceIdentifier).toEqual({ OnlineEvaluationConfigId: 'oec-123' });
+ });
+
+ it('returns empty list when online eval config not found in template', () => {
+ const emptyTemplate: CfnTemplate = {
+ Resources: {
+ IAMRoleLogicalId: {
+ Type: 'AWS::IAM::Role',
+ Properties: {},
+ },
+ },
+ };
+
+ const logicalId = findLogicalIdByProperty(
+ emptyTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'QualityMonitor'
+ );
+
+ expect(logicalId).toBeUndefined();
+ });
+});
+
+// ============================================================================
+// buildImportTemplate Tests for Online Eval Config Resources
+// ============================================================================
+
+describe('buildImportTemplate with Online Eval Config', () => {
+ it('adds online eval config resource to deployed template with Retain deletion policy', () => {
+ const deployedTemplate: CfnTemplate = {
+ Resources: {
+ ExistingIAMRole: {
+ Type: 'AWS::IAM::Role',
+ Properties: { RoleName: 'ExistingRole' },
+ },
+ },
+ };
+
+ const synthTemplate: CfnTemplate = {
+ Resources: {
+ ExistingIAMRole: {
+ Type: 'AWS::IAM::Role',
+ Properties: { RoleName: 'ExistingRole' },
+ },
+ OnlineEvalLogicalId: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'QualityMonitor',
+ },
+ DependsOn: 'ExistingIAMRole',
+ },
+ },
+ };
+
+ const importTemplate = buildImportTemplate(deployedTemplate, synthTemplate, ['OnlineEvalLogicalId']);
+
+ // Verify online eval config resource was added
+ expect(importTemplate.Resources.OnlineEvalLogicalId).toBeDefined();
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.Type).toBe('AWS::BedrockAgentCore::OnlineEvaluationConfig');
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.DeletionPolicy).toBe('Retain');
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.UpdateReplacePolicy).toBe('Retain');
+
+ // DependsOn should be removed for import
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.DependsOn).toBeUndefined();
+
+ // Original resource should still be there
+ expect(importTemplate.Resources.ExistingIAMRole).toBeDefined();
+ });
+});

src/cli/commands/import/command.ts

@@ -1,14 +1,13 @@
 import { handleImport } from './actions';
+import { ANSI } from './constants';
+import { registerImportEvaluator } from './import-evaluator';
 import { registerImportMemory } from './import-memory';
+import { registerImportOnlineEval } from './import-online-eval';
 import { registerImportRuntime } from './import-runtime';
 import type { Command } from '@commander-js/extra-typings';
 import * as fs from 'node:fs';
 
-const green = '\x1b[32m';
-const yellow = '\x1b[33m';
-const cyan = '\x1b[36m';
-const dim = '\x1b[2m';
-const reset = '\x1b[0m';
+const { green, yellow, cyan, dim, reset } = ANSI;
 
 export const registerImport = (program: Command) => {
  const importCmd = program
@@ -151,4 +150,6 @@ export const registerImport = (program: Command) => {
  // Register subcommands for importing individual resource types from AWS
  registerImportRuntime(importCmd);
  registerImportMemory(importCmd);
+ registerImportEvaluator(importCmd);
+ registerImportOnlineEval(importCmd);
 };

src/cli/commands/import/__tests__/import-online-eval.test.ts

@@ -0,0 +1,368 @@
+/**
+ * Import Online Eval Config Unit Tests
+ *
+ * Covers:
+ * - extractAgentName: service name parsing
+ * - toOnlineEvalConfigSpec conversion: happy path, missing sampling, enableOnCreate
+ * - Template logical ID lookup for online eval configs
+ * - Phase 2 import resource list construction for online eval configs
+ */
+import type { GetOnlineEvalConfigResult } from '../../../aws/agentcore-control';
+import { extractAgentName, toOnlineEvalConfigSpec } from '../import-online-eval';
+import { buildImportTemplate, findLogicalIdByProperty, findLogicalIdsByType } from '../template-utils';
+import type { CfnTemplate } from '../template-utils';
+import type { ResourceToImport } from '../types';
+import { describe, expect, it } from 'vitest';
+
+// ============================================================================
+// extractAgentName Tests
+// ============================================================================
+
+describe('extractAgentName', () => {
+ it('extracts agent name from service name with .DEFAULT suffix', () => {
+ expect(extractAgentName(['my_agent.DEFAULT'])).toBe('my_agent');
+ });
+
+ it('extracts agent name with project prefix pattern', () => {
+ expect(extractAgentName(['testproject_my_agent.DEFAULT'])).toBe('testproject_my_agent');
+ });
+
+ it('returns full string when no dot suffix', () => {
+ expect(extractAgentName(['my_agent'])).toBe('my_agent');
+ });
+
+ it('returns undefined for empty array', () => {
+ expect(extractAgentName([])).toBeUndefined();
+ });
+
+ it('uses first service name when multiple provided', () => {
+ expect(extractAgentName(['agent_one.DEFAULT', 'agent_two.DEFAULT'])).toBe('agent_one');
+ });
+
+ it('handles service name with multiple dots', () => {
+ expect(extractAgentName(['my.agent.DEFAULT'])).toBe('my.agent');
+ });
+});
+
+// ============================================================================
+// toOnlineEvalConfigSpec Conversion Tests
+// ============================================================================
+
+describe('toOnlineEvalConfigSpec', () => {
+ it('maps online eval config with all fields', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-123',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-123',
+ configName: 'QualityMonitor',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ description: 'Monitor agent quality',
+ samplingPercentage: 50,
+ serviceNames: ['my_agent.DEFAULT'],
+ evaluatorIds: ['eval-456'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'QualityMonitor', 'my_agent', ['my_evaluator']);
+
+ expect(result.name).toBe('QualityMonitor');
+ expect(result.agent).toBe('my_agent');
+ expect(result.evaluators).toEqual(['my_evaluator']);
+ expect(result.samplingRate).toBe(50);
+ expect(result.description).toBe('Monitor agent quality');
+ expect(result.enableOnCreate).toBe(true);
+ });
+
+ it('omits enableOnCreate when execution status is DISABLED', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-456',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-456',
+ configName: 'DisabledConfig',
+ status: 'ACTIVE',
+ executionStatus: 'DISABLED',
+ samplingPercentage: 10,
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'DisabledConfig', 'agent', ['eval_one']);
+
+ expect(result.enableOnCreate).toBeUndefined();
+ });
+
+ it('omits description when not present', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-789',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-789',
+ configName: 'NoDesc',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ samplingPercentage: 25,
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'NoDesc', 'agent', ['eval_one']);
+
+ expect(result.description).toBeUndefined();
+ });
+
+ it('throws when sampling percentage is missing', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-no-sampling',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-no-sampling',
+ configName: 'NoSampling',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1'],
+ };
+
+ expect(() => toOnlineEvalConfigSpec(detail, 'NoSampling', 'agent', ['eval_one'])).toThrow(
+ 'has no sampling configuration'
+ );
+ });
+
+ it('supports multiple evaluator references', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-multi',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-multi',
+ configName: 'MultiEval',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ samplingPercentage: 75,
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1', 'eval-2'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'MultiEval', 'agent', [
+ 'local_eval',
+ 'arn:aws:bedrock-agentcore:us-west-2:123456789012:evaluator/eval-2',
+ ]);
+
+ expect(result.evaluators).toHaveLength(2);
+ expect(result.evaluators[0]).toBe('local_eval');
+ expect(result.evaluators[1]).toMatch(/^arn:/);
+ });
+});
+
+// ============================================================================
+// Template Logical ID Lookup Tests for Online Eval Configs
+// ============================================================================
+
+describe('Template Logical ID Lookup for Online Eval Configs', () => {
+ const synthTemplate: CfnTemplate = {
+ Resources: {
+ MyOnlineEvalConfig: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'QualityMonitor',
+ },
+ },
+ PrefixedOnlineEvalConfig: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'TestProject_PrefixedConfig',
+ },
+ },
+ MyAgentRuntime: {
+ Type: 'AWS::BedrockAgentCore::Runtime',
+ Properties: {
+ AgentRuntimeName: 'TestProject_my_agent',
+ },
+ },
+ MyIAMRole: {
+ Type: 'AWS::IAM::Role',
+ Properties: {
+ RoleName: 'MyRole',
+ },
+ },
+ },
+ };
+
+ it('finds online eval config logical ID by OnlineEvaluationConfigName property', () => {
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'QualityMonitor'
+ );
+ expect(logicalId).toBe('MyOnlineEvalConfig');
+ });
+
+ it('finds prefixed online eval config by full name', () => {
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'TestProject_PrefixedConfig'
+ );
+ expect(logicalId).toBe('PrefixedOnlineEvalConfig');
+ });
+
+ it('finds all online eval config logical IDs by type', () => {
+ const logicalIds = findLogicalIdsByType(synthTemplate, 'AWS::BedrockAgentCore::OnlineEvaluationConfig');
+ expect(logicalIds).toHaveLength(2);
+ expect(logicalIds).toContain('MyOnlineEvalConfig');
+ expect(logicalIds).toContain('PrefixedOnlineEvalConfig');
+ });
+
+ it('returns undefined for non-existent config name', () => {
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'nonexistent_config'
+ );
+ expect(logicalId).toBeUndefined();
+ });
+
+ it('falls back to single online eval config logical ID when name does not match', () => {
+ const singleConfigTemplate: CfnTemplate = {
+ Resources: {
+ OnlyConfig: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'some_config',
+ },
+ },
+ },
+ };
+
+ let logicalId = findLogicalIdByProperty(
+ singleConfigTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'different_name'
+ );
+
+ // Primary lookup fails
+ expect(logicalId).toBeUndefined();
+
+ // Fallback: if there's only one config resource, use it
+ if (!logicalId) {
+ const configLogicalIds = findLogicalIdsByType(
+ singleConfigTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig'
+ );
+ if (configLogicalIds.length === 1) {
+ logicalId = configLogicalIds[0];
+ }
+ }
+ expect(logicalId).toBe('OnlyConfig');
+ });
+});
+
+// ============================================================================
+// Phase 2 Resource Import List Construction for Online Eval Configs
+// ============================================================================
+
+describe('Phase 2: ResourceToImport List Construction for Online Eval Configs', () => {
+ const synthTemplate: CfnTemplate = {
+ Resources: {
+ OnlineEvalLogicalId: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'QualityMonitor',
+ },
+ },
+ IAMRoleLogicalId: {
+ Type: 'AWS::IAM::Role',
+ Properties: {},
+ },
+ },
+ };
+
+ it('builds ResourceToImport list for online eval config', () => {
+ const configName = 'QualityMonitor';
+ const configId = 'oec-123';
+
+ const resourcesToImport: ResourceToImport[] = [];
+
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ configName
+ );
+
+ if (logicalId) {
+ resourcesToImport.push({
+ resourceType: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ logicalResourceId: logicalId,
+ resourceIdentifier: { OnlineEvaluationConfigId: configId },
+ });
+ }
+
+ expect(resourcesToImport).toHaveLength(1);
+ expect(resourcesToImport[0]!.resourceType).toBe('AWS::BedrockAgentCore::OnlineEvaluationConfig');
+ expect(resourcesToImport[0]!.logicalResourceId).toBe('OnlineEvalLogicalId');
+ expect(resourcesToImport[0]!.resourceIdentifier).toEqual({ OnlineEvaluationConfigId: 'oec-123' });
+ });
+
+ it('returns empty list when online eval config not found in template', () => {
+ const emptyTemplate: CfnTemplate = {
+ Resources: {
+ IAMRoleLogicalId: {
+ Type: 'AWS::IAM::Role',
+ Properties: {},
+ },
+ },
+ };
+
+ const logicalId = findLogicalIdByProperty(
+ emptyTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'QualityMonitor'
+ );
+
+ expect(logicalId).toBeUndefined();
+ });
+});
+
+// ============================================================================
+// buildImportTemplate Tests for Online Eval Config Resources
+// ============================================================================
+
+describe('buildImportTemplate with Online Eval Config', () => {
+ it('adds online eval config resource to deployed template with Retain deletion policy', () => {
+ const deployedTemplate: CfnTemplate = {
+ Resources: {
+ ExistingIAMRole: {
+ Type: 'AWS::IAM::Role',
+ Properties: { RoleName: 'ExistingRole' },
+ },
+ },
+ };
+
+ const synthTemplate: CfnTemplate = {
+ Resources: {
+ ExistingIAMRole: {
+ Type: 'AWS::IAM::Role',
+ Properties: { RoleName: 'ExistingRole' },
+ },
+ OnlineEvalLogicalId: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'QualityMonitor',
+ },
+ DependsOn: 'ExistingIAMRole',
+ },
+ },
+ };
+
+ const importTemplate = buildImportTemplate(deployedTemplate, synthTemplate, ['OnlineEvalLogicalId']);
+
+ // Verify online eval config resource was added
+ expect(importTemplate.Resources.OnlineEvalLogicalId).toBeDefined();
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.Type).toBe('AWS::BedrockAgentCore::OnlineEvaluationConfig');
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.DeletionPolicy).toBe('Retain');
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.UpdateReplacePolicy).toBe('Retain');
+
+ // DependsOn should be removed for import
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.DependsOn).toBeUndefined();
+
+ // Original resource should still be there
+ expect(importTemplate.Resources.ExistingIAMRole).toBeDefined();
+ });
+});

src/cli/commands/import/command.ts

@@ -1,14 +1,13 @@
 import { handleImport } from './actions';
+import { ANSI } from './constants';
+import { registerImportEvaluator } from './import-evaluator';
 import { registerImportMemory } from './import-memory';
+import { registerImportOnlineEval } from './import-online-eval';
 import { registerImportRuntime } from './import-runtime';
 import type { Command } from '@commander-js/extra-typings';
 import * as fs from 'node:fs';
 
-const green = '\x1b[32m';
-const yellow = '\x1b[33m';
-const cyan = '\x1b[36m';
-const dim = '\x1b[2m';
-const reset = '\x1b[0m';
+const { green, yellow, cyan, dim, reset } = ANSI;
 
 export const registerImport = (program: Command) => {
  const importCmd = program
@@ -151,4 +150,6 @@ export const registerImport = (program: Command) => {
  // Register subcommands for importing individual resource types from AWS
  registerImportRuntime(importCmd);
  registerImportMemory(importCmd);
+ registerImportEvaluator(importCmd);
+ registerImportOnlineEval(importCmd);
 };

src/cli/commands/import/__tests__/import-online-eval.test.ts

@@ -0,0 +1,368 @@
+/**
+ * Import Online Eval Config Unit Tests
+ *
+ * Covers:
+ * - extractAgentName: service name parsing
+ * - toOnlineEvalConfigSpec conversion: happy path, missing sampling, enableOnCreate
+ * - Template logical ID lookup for online eval configs
+ * - Phase 2 import resource list construction for online eval configs
+ */
+import type { GetOnlineEvalConfigResult } from '../../../aws/agentcore-control';
+import { extractAgentName, toOnlineEvalConfigSpec } from '../import-online-eval';
+import { buildImportTemplate, findLogicalIdByProperty, findLogicalIdsByType } from '../template-utils';
+import type { CfnTemplate } from '../template-utils';
+import type { ResourceToImport } from '../types';
+import { describe, expect, it } from 'vitest';
+
+// ============================================================================
+// extractAgentName Tests
+// ============================================================================
+
+describe('extractAgentName', () => {
+ it('extracts agent name from service name with .DEFAULT suffix', () => {
+ expect(extractAgentName(['my_agent.DEFAULT'])).toBe('my_agent');
+ });
+
+ it('extracts agent name with project prefix pattern', () => {
+ expect(extractAgentName(['testproject_my_agent.DEFAULT'])).toBe('testproject_my_agent');
+ });
+
+ it('returns full string when no dot suffix', () => {
+ expect(extractAgentName(['my_agent'])).toBe('my_agent');
+ });
+
+ it('returns undefined for empty array', () => {
+ expect(extractAgentName([])).toBeUndefined();
+ });
+
+ it('uses first service name when multiple provided', () => {
+ expect(extractAgentName(['agent_one.DEFAULT', 'agent_two.DEFAULT'])).toBe('agent_one');
+ });
+
+ it('handles service name with multiple dots', () => {
+ expect(extractAgentName(['my.agent.DEFAULT'])).toBe('my.agent');
+ });
+});
+
+// ============================================================================
+// toOnlineEvalConfigSpec Conversion Tests
+// ============================================================================
+
+describe('toOnlineEvalConfigSpec', () => {
+ it('maps online eval config with all fields', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-123',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-123',
+ configName: 'QualityMonitor',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ description: 'Monitor agent quality',
+ samplingPercentage: 50,
+ serviceNames: ['my_agent.DEFAULT'],
+ evaluatorIds: ['eval-456'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'QualityMonitor', 'my_agent', ['my_evaluator']);
+
+ expect(result.name).toBe('QualityMonitor');
+ expect(result.agent).toBe('my_agent');
+ expect(result.evaluators).toEqual(['my_evaluator']);
+ expect(result.samplingRate).toBe(50);
+ expect(result.description).toBe('Monitor agent quality');
+ expect(result.enableOnCreate).toBe(true);
+ });
+
+ it('omits enableOnCreate when execution status is DISABLED', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-456',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-456',
+ configName: 'DisabledConfig',
+ status: 'ACTIVE',
+ executionStatus: 'DISABLED',
+ samplingPercentage: 10,
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'DisabledConfig', 'agent', ['eval_one']);
+
+ expect(result.enableOnCreate).toBeUndefined();
+ });
+
+ it('omits description when not present', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-789',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-789',
+ configName: 'NoDesc',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ samplingPercentage: 25,
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'NoDesc', 'agent', ['eval_one']);
+
+ expect(result.description).toBeUndefined();
+ });
+
+ it('throws when sampling percentage is missing', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-no-sampling',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-no-sampling',
+ configName: 'NoSampling',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1'],
+ };
+
+ expect(() => toOnlineEvalConfigSpec(detail, 'NoSampling', 'agent', ['eval_one'])).toThrow(
+ 'has no sampling configuration'
+ );
+ });
+
+ it('supports multiple evaluator references', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-multi',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-multi',
+ configName: 'MultiEval',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ samplingPercentage: 75,
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1', 'eval-2'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'MultiEval', 'agent', [
+ 'local_eval',
+ 'arn:aws:bedrock-agentcore:us-west-2:123456789012:evaluator/eval-2',
+ ]);
+
+ expect(result.evaluators).toHaveLength(2);
+ expect(result.evaluators[0]).toBe('local_eval');
+ expect(result.evaluators[1]).toMatch(/^arn:/);
+ });
+});
+
+// ============================================================================
+// Template Logical ID Lookup Tests for Online Eval Configs
+// ============================================================================
+
+describe('Template Logical ID Lookup for Online Eval Configs', () => {
+ const synthTemplate: CfnTemplate = {
+ Resources: {
+ MyOnlineEvalConfig: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'QualityMonitor',
+ },
+ },
+ PrefixedOnlineEvalConfig: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'TestProject_PrefixedConfig',
+ },
+ },
+ MyAgentRuntime: {
+ Type: 'AWS::BedrockAgentCore::Runtime',
+ Properties: {
+ AgentRuntimeName: 'TestProject_my_agent',
+ },
+ },
+ MyIAMRole: {
+ Type: 'AWS::IAM::Role',
+ Properties: {
+ RoleName: 'MyRole',
+ },
+ },
+ },
+ };
+
+ it('finds online eval config logical ID by OnlineEvaluationConfigName property', () => {
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'QualityMonitor'
+ );
+ expect(logicalId).toBe('MyOnlineEvalConfig');
+ });
+
+ it('finds prefixed online eval config by full name', () => {
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'TestProject_PrefixedConfig'
+ );
+ expect(logicalId).toBe('PrefixedOnlineEvalConfig');
+ });
+
+ it('finds all online eval config logical IDs by type', () => {
+ const logicalIds = findLogicalIdsByType(synthTemplate, 'AWS::BedrockAgentCore::OnlineEvaluationConfig');
+ expect(logicalIds).toHaveLength(2);
+ expect(logicalIds).toContain('MyOnlineEvalConfig');
+ expect(logicalIds).toContain('PrefixedOnlineEvalConfig');
+ });
+
+ it('returns undefined for non-existent config name', () => {
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'nonexistent_config'
+ );
+ expect(logicalId).toBeUndefined();
+ });
+
+ it('falls back to single online eval config logical ID when name does not match', () => {
+ const singleConfigTemplate: CfnTemplate = {
+ Resources: {
+ OnlyConfig: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'some_config',
+ },
+ },
+ },
+ };
+
+ let logicalId = findLogicalIdByProperty(
+ singleConfigTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'different_name'
+ );
+
+ // Primary lookup fails
+ expect(logicalId).toBeUndefined();
+
+ // Fallback: if there's only one config resource, use it
+ if (!logicalId) {
+ const configLogicalIds = findLogicalIdsByType(
+ singleConfigTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig'
+ );
+ if (configLogicalIds.length === 1) {
+ logicalId = configLogicalIds[0];
+ }
+ }
+ expect(logicalId).toBe('OnlyConfig');
+ });
+});
+
+// ============================================================================
+// Phase 2 Resource Import List Construction for Online Eval Configs
+// ============================================================================
+
+describe('Phase 2: ResourceToImport List Construction for Online Eval Configs', () => {
+ const synthTemplate: CfnTemplate = {
+ Resources: {
+ OnlineEvalLogicalId: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'QualityMonitor',
+ },
+ },
+ IAMRoleLogicalId: {
+ Type: 'AWS::IAM::Role',
+ Properties: {},
+ },
+ },
+ };
+
+ it('builds ResourceToImport list for online eval config', () => {
+ const configName = 'QualityMonitor';
+ const configId = 'oec-123';
+
+ const resourcesToImport: ResourceToImport[] = [];
+
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ configName
+ );
+
+ if (logicalId) {
+ resourcesToImport.push({
+ resourceType: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ logicalResourceId: logicalId,
+ resourceIdentifier: { OnlineEvaluationConfigId: configId },
+ });
+ }
+
+ expect(resourcesToImport).toHaveLength(1);
+ expect(resourcesToImport[0]!.resourceType).toBe('AWS::BedrockAgentCore::OnlineEvaluationConfig');
+ expect(resourcesToImport[0]!.logicalResourceId).toBe('OnlineEvalLogicalId');
+ expect(resourcesToImport[0]!.resourceIdentifier).toEqual({ OnlineEvaluationConfigId: 'oec-123' });
+ });
+
+ it('returns empty list when online eval config not found in template', () => {
+ const emptyTemplate: CfnTemplate = {
+ Resources: {
+ IAMRoleLogicalId: {
+ Type: 'AWS::IAM::Role',
+ Properties: {},
+ },
+ },
+ };
+
+ const logicalId = findLogicalIdByProperty(
+ emptyTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'QualityMonitor'
+ );
+
+ expect(logicalId).toBeUndefined();
+ });
+});
+
+// ============================================================================
+// buildImportTemplate Tests for Online Eval Config Resources
+// ============================================================================
+
+describe('buildImportTemplate with Online Eval Config', () => {
+ it('adds online eval config resource to deployed template with Retain deletion policy', () => {
+ const deployedTemplate: CfnTemplate = {
+ Resources: {
+ ExistingIAMRole: {
+ Type: 'AWS::IAM::Role',
+ Properties: { RoleName: 'ExistingRole' },
+ },
+ },
+ };
+
+ const synthTemplate: CfnTemplate = {
+ Resources: {
+ ExistingIAMRole: {
+ Type: 'AWS::IAM::Role',
+ Properties: { RoleName: 'ExistingRole' },
+ },
+ OnlineEvalLogicalId: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'QualityMonitor',
+ },
+ DependsOn: 'ExistingIAMRole',
+ },
+ },
+ };
+
+ const importTemplate = buildImportTemplate(deployedTemplate, synthTemplate, ['OnlineEvalLogicalId']);
+
+ // Verify online eval config resource was added
+ expect(importTemplate.Resources.OnlineEvalLogicalId).toBeDefined();
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.Type).toBe('AWS::BedrockAgentCore::OnlineEvaluationConfig');
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.DeletionPolicy).toBe('Retain');
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.UpdateReplacePolicy).toBe('Retain');
+
+ // DependsOn should be removed for import
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.DependsOn).toBeUndefined();
+
+ // Original resource should still be there
+ expect(importTemplate.Resources.ExistingIAMRole).toBeDefined();
+ });
+});

src/cli/commands/import/command.ts

@@ -1,14 +1,13 @@
 import { handleImport } from './actions';
+import { ANSI } from './constants';
+import { registerImportEvaluator } from './import-evaluator';
 import { registerImportMemory } from './import-memory';
+import { registerImportOnlineEval } from './import-online-eval';
 import { registerImportRuntime } from './import-runtime';
 import type { Command } from '@commander-js/extra-typings';
 import * as fs from 'node:fs';
 
-const green = '\x1b[32m';
-const yellow = '\x1b[33m';
-const cyan = '\x1b[36m';
-const dim = '\x1b[2m';
-const reset = '\x1b[0m';
+const { green, yellow, cyan, dim, reset } = ANSI;
 
 export const registerImport = (program: Command) => {
  const importCmd = program
@@ -151,4 +150,6 @@ export const registerImport = (program: Command) => {
  // Register subcommands for importing individual resource types from AWS
  registerImportRuntime(importCmd);
  registerImportMemory(importCmd);
+ registerImportEvaluator(importCmd);
+ registerImportOnlineEval(importCmd);
 };

src/cli/commands/import/__tests__/import-online-eval.test.ts

@@ -0,0 +1,368 @@
+/**
+ * Import Online Eval Config Unit Tests
+ *
+ * Covers:
+ * - extractAgentName: service name parsing
+ * - toOnlineEvalConfigSpec conversion: happy path, missing sampling, enableOnCreate
+ * - Template logical ID lookup for online eval configs
+ * - Phase 2 import resource list construction for online eval configs
+ */
+import type { GetOnlineEvalConfigResult } from '../../../aws/agentcore-control';
+import { extractAgentName, toOnlineEvalConfigSpec } from '../import-online-eval';
+import { buildImportTemplate, findLogicalIdByProperty, findLogicalIdsByType } from '../template-utils';
+import type { CfnTemplate } from '../template-utils';
+import type { ResourceToImport } from '../types';
+import { describe, expect, it } from 'vitest';
+
+// ============================================================================
+// extractAgentName Tests
+// ============================================================================
+
+describe('extractAgentName', () => {
+ it('extracts agent name from service name with .DEFAULT suffix', () => {
+ expect(extractAgentName(['my_agent.DEFAULT'])).toBe('my_agent');
+ });
+
+ it('extracts agent name with project prefix pattern', () => {
+ expect(extractAgentName(['testproject_my_agent.DEFAULT'])).toBe('testproject_my_agent');
+ });
+
+ it('returns full string when no dot suffix', () => {
+ expect(extractAgentName(['my_agent'])).toBe('my_agent');
+ });
+
+ it('returns undefined for empty array', () => {
+ expect(extractAgentName([])).toBeUndefined();
+ });
+
+ it('uses first service name when multiple provided', () => {
+ expect(extractAgentName(['agent_one.DEFAULT', 'agent_two.DEFAULT'])).toBe('agent_one');
+ });
+
+ it('handles service name with multiple dots', () => {
+ expect(extractAgentName(['my.agent.DEFAULT'])).toBe('my.agent');
+ });
+});
+
+// ============================================================================
+// toOnlineEvalConfigSpec Conversion Tests
+// ============================================================================
+
+describe('toOnlineEvalConfigSpec', () => {
+ it('maps online eval config with all fields', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-123',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-123',
+ configName: 'QualityMonitor',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ description: 'Monitor agent quality',
+ samplingPercentage: 50,
+ serviceNames: ['my_agent.DEFAULT'],
+ evaluatorIds: ['eval-456'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'QualityMonitor', 'my_agent', ['my_evaluator']);
+
+ expect(result.name).toBe('QualityMonitor');
+ expect(result.agent).toBe('my_agent');
+ expect(result.evaluators).toEqual(['my_evaluator']);
+ expect(result.samplingRate).toBe(50);
+ expect(result.description).toBe('Monitor agent quality');
+ expect(result.enableOnCreate).toBe(true);
+ });
+
+ it('omits enableOnCreate when execution status is DISABLED', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-456',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-456',
+ configName: 'DisabledConfig',
+ status: 'ACTIVE',
+ executionStatus: 'DISABLED',
+ samplingPercentage: 10,
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'DisabledConfig', 'agent', ['eval_one']);
+
+ expect(result.enableOnCreate).toBeUndefined();
+ });
+
+ it('omits description when not present', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-789',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-789',
+ configName: 'NoDesc',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ samplingPercentage: 25,
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'NoDesc', 'agent', ['eval_one']);
+
+ expect(result.description).toBeUndefined();
+ });
+
+ it('throws when sampling percentage is missing', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-no-sampling',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-no-sampling',
+ configName: 'NoSampling',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1'],
+ };
+
+ expect(() => toOnlineEvalConfigSpec(detail, 'NoSampling', 'agent', ['eval_one'])).toThrow(
+ 'has no sampling configuration'
+ );
+ });
+
+ it('supports multiple evaluator references', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-multi',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-multi',
+ configName: 'MultiEval',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ samplingPercentage: 75,
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1', 'eval-2'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'MultiEval', 'agent', [
+ 'local_eval',
+ 'arn:aws:bedrock-agentcore:us-west-2:123456789012:evaluator/eval-2',
+ ]);
+
+ expect(result.evaluators).toHaveLength(2);
+ expect(result.evaluators[0]).toBe('local_eval');
+ expect(result.evaluators[1]).toMatch(/^arn:/);
+ });
+});
+
+// ============================================================================
+// Template Logical ID Lookup Tests for Online Eval Configs
+// ============================================================================
+
+describe('Template Logical ID Lookup for Online Eval Configs', () => {
+ const synthTemplate: CfnTemplate = {
+ Resources: {
+ MyOnlineEvalConfig: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'QualityMonitor',
+ },
+ },
+ PrefixedOnlineEvalConfig: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'TestProject_PrefixedConfig',
+ },
+ },
+ MyAgentRuntime: {
+ Type: 'AWS::BedrockAgentCore::Runtime',
+ Properties: {
+ AgentRuntimeName: 'TestProject_my_agent',
+ },
+ },
+ MyIAMRole: {
+ Type: 'AWS::IAM::Role',
+ Properties: {
+ RoleName: 'MyRole',
+ },
+ },
+ },
+ };
+
+ it('finds online eval config logical ID by OnlineEvaluationConfigName property', () => {
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'QualityMonitor'
+ );
+ expect(logicalId).toBe('MyOnlineEvalConfig');
+ });
+
+ it('finds prefixed online eval config by full name', () => {
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'TestProject_PrefixedConfig'
+ );
+ expect(logicalId).toBe('PrefixedOnlineEvalConfig');
+ });
+
+ it('finds all online eval config logical IDs by type', () => {
+ const logicalIds = findLogicalIdsByType(synthTemplate, 'AWS::BedrockAgentCore::OnlineEvaluationConfig');
+ expect(logicalIds).toHaveLength(2);
+ expect(logicalIds).toContain('MyOnlineEvalConfig');
+ expect(logicalIds).toContain('PrefixedOnlineEvalConfig');
+ });
+
+ it('returns undefined for non-existent config name', () => {
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'nonexistent_config'
+ );
+ expect(logicalId).toBeUndefined();
+ });
+
+ it('falls back to single online eval config logical ID when name does not match', () => {
+ const singleConfigTemplate: CfnTemplate = {
+ Resources: {
+ OnlyConfig: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'some_config',
+ },
+ },
+ },
+ };
+
+ let logicalId = findLogicalIdByProperty(
+ singleConfigTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'different_name'
+ );
+
+ // Primary lookup fails
+ expect(logicalId).toBeUndefined();
+
+ // Fallback: if there's only one config resource, use it
+ if (!logicalId) {
+ const configLogicalIds = findLogicalIdsByType(
+ singleConfigTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig'
+ );
+ if (configLogicalIds.length === 1) {
+ logicalId = configLogicalIds[0];
+ }
+ }
+ expect(logicalId).toBe('OnlyConfig');
+ });
+});
+
+// ============================================================================
+// Phase 2 Resource Import List Construction for Online Eval Configs
+// ============================================================================
+
+describe('Phase 2: ResourceToImport List Construction for Online Eval Configs', () => {
+ const synthTemplate: CfnTemplate = {
+ Resources: {
+ OnlineEvalLogicalId: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'QualityMonitor',
+ },
+ },
+ IAMRoleLogicalId: {
+ Type: 'AWS::IAM::Role',
+ Properties: {},
+ },
+ },
+ };
+
+ it('builds ResourceToImport list for online eval config', () => {
+ const configName = 'QualityMonitor';
+ const configId = 'oec-123';
+
+ const resourcesToImport: ResourceToImport[] = [];
+
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ configName
+ );
+
+ if (logicalId) {
+ resourcesToImport.push({
+ resourceType: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ logicalResourceId: logicalId,
+ resourceIdentifier: { OnlineEvaluationConfigId: configId },
+ });
+ }
+
+ expect(resourcesToImport).toHaveLength(1);
+ expect(resourcesToImport[0]!.resourceType).toBe('AWS::BedrockAgentCore::OnlineEvaluationConfig');
+ expect(resourcesToImport[0]!.logicalResourceId).toBe('OnlineEvalLogicalId');
+ expect(resourcesToImport[0]!.resourceIdentifier).toEqual({ OnlineEvaluationConfigId: 'oec-123' });
+ });
+
+ it('returns empty list when online eval config not found in template', () => {
+ const emptyTemplate: CfnTemplate = {
+ Resources: {
+ IAMRoleLogicalId: {
+ Type: 'AWS::IAM::Role',
+ Properties: {},
+ },
+ },
+ };
+
+ const logicalId = findLogicalIdByProperty(
+ emptyTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'QualityMonitor'
+ );
+
+ expect(logicalId).toBeUndefined();
+ });
+});
+
+// ============================================================================
+// buildImportTemplate Tests for Online Eval Config Resources
+// ============================================================================
+
+describe('buildImportTemplate with Online Eval Config', () => {
+ it('adds online eval config resource to deployed template with Retain deletion policy', () => {
+ const deployedTemplate: CfnTemplate = {
+ Resources: {
+ ExistingIAMRole: {
+ Type: 'AWS::IAM::Role',
+ Properties: { RoleName: 'ExistingRole' },
+ },
+ },
+ };
+
+ const synthTemplate: CfnTemplate = {
+ Resources: {
+ ExistingIAMRole: {
+ Type: 'AWS::IAM::Role',
+ Properties: { RoleName: 'ExistingRole' },
+ },
+ OnlineEvalLogicalId: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'QualityMonitor',
+ },
+ DependsOn: 'ExistingIAMRole',
+ },
+ },
+ };
+
+ const importTemplate = buildImportTemplate(deployedTemplate, synthTemplate, ['OnlineEvalLogicalId']);
+
+ // Verify online eval config resource was added
+ expect(importTemplate.Resources.OnlineEvalLogicalId).toBeDefined();
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.Type).toBe('AWS::BedrockAgentCore::OnlineEvaluationConfig');
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.DeletionPolicy).toBe('Retain');
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.UpdateReplacePolicy).toBe('Retain');
+
+ // DependsOn should be removed for import
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.DependsOn).toBeUndefined();
+
+ // Original resource should still be there
+ expect(importTemplate.Resources.ExistingIAMRole).toBeDefined();
+ });
+});

src/cli/commands/import/command.ts

@@ -1,14 +1,13 @@
 import { handleImport } from './actions';
+import { ANSI } from './constants';
+import { registerImportEvaluator } from './import-evaluator';
 import { registerImportMemory } from './import-memory';
+import { registerImportOnlineEval } from './import-online-eval';
 import { registerImportRuntime } from './import-runtime';
 import type { Command } from '@commander-js/extra-typings';
 import * as fs from 'node:fs';
 
-const green = '\x1b[32m';
-const yellow = '\x1b[33m';
-const cyan = '\x1b[36m';
-const dim = '\x1b[2m';
-const reset = '\x1b[0m';
+const { green, yellow, cyan, dim, reset } = ANSI;
 
 export const registerImport = (program: Command) => {
  const importCmd = program
@@ -151,4 +150,6 @@ export const registerImport = (program: Command) => {
  // Register subcommands for importing individual resource types from AWS
  registerImportRuntime(importCmd);
  registerImportMemory(importCmd);
+ registerImportEvaluator(importCmd);
+ registerImportOnlineEval(importCmd);
 };

src/cli/commands/import/__tests__/import-online-eval.test.ts

@@ -0,0 +1,368 @@
+/**
+ * Import Online Eval Config Unit Tests
+ *
+ * Covers:
+ * - extractAgentName: service name parsing
+ * - toOnlineEvalConfigSpec conversion: happy path, missing sampling, enableOnCreate
+ * - Template logical ID lookup for online eval configs
+ * - Phase 2 import resource list construction for online eval configs
+ */
+import type { GetOnlineEvalConfigResult } from '../../../aws/agentcore-control';
+import { extractAgentName, toOnlineEvalConfigSpec } from '../import-online-eval';
+import { buildImportTemplate, findLogicalIdByProperty, findLogicalIdsByType } from '../template-utils';
+import type { CfnTemplate } from '../template-utils';
+import type { ResourceToImport } from '../types';
+import { describe, expect, it } from 'vitest';
+
+// ============================================================================
+// extractAgentName Tests
+// ============================================================================
+
+describe('extractAgentName', () => {
+ it('extracts agent name from service name with .DEFAULT suffix', () => {
+ expect(extractAgentName(['my_agent.DEFAULT'])).toBe('my_agent');
+ });
+
+ it('extracts agent name with project prefix pattern', () => {
+ expect(extractAgentName(['testproject_my_agent.DEFAULT'])).toBe('testproject_my_agent');
+ });
+
+ it('returns full string when no dot suffix', () => {
+ expect(extractAgentName(['my_agent'])).toBe('my_agent');
+ });
+
+ it('returns undefined for empty array', () => {
+ expect(extractAgentName([])).toBeUndefined();
+ });
+
+ it('uses first service name when multiple provided', () => {
+ expect(extractAgentName(['agent_one.DEFAULT', 'agent_two.DEFAULT'])).toBe('agent_one');
+ });
+
+ it('handles service name with multiple dots', () => {
+ expect(extractAgentName(['my.agent.DEFAULT'])).toBe('my.agent');
+ });
+});
+
+// ============================================================================
+// toOnlineEvalConfigSpec Conversion Tests
+// ============================================================================
+
+describe('toOnlineEvalConfigSpec', () => {
+ it('maps online eval config with all fields', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-123',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-123',
+ configName: 'QualityMonitor',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ description: 'Monitor agent quality',
+ samplingPercentage: 50,
+ serviceNames: ['my_agent.DEFAULT'],
+ evaluatorIds: ['eval-456'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'QualityMonitor', 'my_agent', ['my_evaluator']);
+
+ expect(result.name).toBe('QualityMonitor');
+ expect(result.agent).toBe('my_agent');
+ expect(result.evaluators).toEqual(['my_evaluator']);
+ expect(result.samplingRate).toBe(50);
+ expect(result.description).toBe('Monitor agent quality');
+ expect(result.enableOnCreate).toBe(true);
+ });
+
+ it('omits enableOnCreate when execution status is DISABLED', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-456',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-456',
+ configName: 'DisabledConfig',
+ status: 'ACTIVE',
+ executionStatus: 'DISABLED',
+ samplingPercentage: 10,
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'DisabledConfig', 'agent', ['eval_one']);
+
+ expect(result.enableOnCreate).toBeUndefined();
+ });
+
+ it('omits description when not present', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-789',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-789',
+ configName: 'NoDesc',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ samplingPercentage: 25,
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'NoDesc', 'agent', ['eval_one']);
+
+ expect(result.description).toBeUndefined();
+ });
+
+ it('throws when sampling percentage is missing', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-no-sampling',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-no-sampling',
+ configName: 'NoSampling',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1'],
+ };
+
+ expect(() => toOnlineEvalConfigSpec(detail, 'NoSampling', 'agent', ['eval_one'])).toThrow(
+ 'has no sampling configuration'
+ );
+ });
+
+ it('supports multiple evaluator references', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-multi',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-multi',
+ configName: 'MultiEval',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ samplingPercentage: 75,
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1', 'eval-2'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'MultiEval', 'agent', [
+ 'local_eval',
+ 'arn:aws:bedrock-agentcore:us-west-2:123456789012:evaluator/eval-2',
+ ]);
+
+ expect(result.evaluators).toHaveLength(2);
+ expect(result.evaluators[0]).toBe('local_eval');
+ expect(result.evaluators[1]).toMatch(/^arn:/);
+ });
+});
+
+// ============================================================================
+// Template Logical ID Lookup Tests for Online Eval Configs
+// ============================================================================
+
+describe('Template Logical ID Lookup for Online Eval Configs', () => {
+ const synthTemplate: CfnTemplate = {
+ Resources: {
+ MyOnlineEvalConfig: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'QualityMonitor',
+ },
+ },
+ PrefixedOnlineEvalConfig: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'TestProject_PrefixedConfig',
+ },
+ },
+ MyAgentRuntime: {
+ Type: 'AWS::BedrockAgentCore::Runtime',
+ Properties: {
+ AgentRuntimeName: 'TestProject_my_agent',
+ },
+ },
+ MyIAMRole: {
+ Type: 'AWS::IAM::Role',
+ Properties: {
+ RoleName: 'MyRole',
+ },
+ },
+ },
+ };
+
+ it('finds online eval config logical ID by OnlineEvaluationConfigName property', () => {
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'QualityMonitor'
+ );
+ expect(logicalId).toBe('MyOnlineEvalConfig');
+ });
+
+ it('finds prefixed online eval config by full name', () => {
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'TestProject_PrefixedConfig'
+ );
+ expect(logicalId).toBe('PrefixedOnlineEvalConfig');
+ });
+
+ it('finds all online eval config logical IDs by type', () => {
+ const logicalIds = findLogicalIdsByType(synthTemplate, 'AWS::BedrockAgentCore::OnlineEvaluationConfig');
+ expect(logicalIds).toHaveLength(2);
+ expect(logicalIds).toContain('MyOnlineEvalConfig');
+ expect(logicalIds).toContain('PrefixedOnlineEvalConfig');
+ });
+
+ it('returns undefined for non-existent config name', () => {
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'nonexistent_config'
+ );
+ expect(logicalId).toBeUndefined();
+ });
+
+ it('falls back to single online eval config logical ID when name does not match', () => {
+ const singleConfigTemplate: CfnTemplate = {
+ Resources: {
+ OnlyConfig: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'some_config',
+ },
+ },
+ },
+ };
+
+ let logicalId = findLogicalIdByProperty(
+ singleConfigTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'different_name'
+ );
+
+ // Primary lookup fails
+ expect(logicalId).toBeUndefined();
+
+ // Fallback: if there's only one config resource, use it
+ if (!logicalId) {
+ const configLogicalIds = findLogicalIdsByType(
+ singleConfigTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig'
+ );
+ if (configLogicalIds.length === 1) {
+ logicalId = configLogicalIds[0];
+ }
+ }
+ expect(logicalId).toBe('OnlyConfig');
+ });
+});
+
+// ============================================================================
+// Phase 2 Resource Import List Construction for Online Eval Configs
+// ============================================================================
+
+describe('Phase 2: ResourceToImport List Construction for Online Eval Configs', () => {
+ const synthTemplate: CfnTemplate = {
+ Resources: {
+ OnlineEvalLogicalId: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'QualityMonitor',
+ },
+ },
+ IAMRoleLogicalId: {
+ Type: 'AWS::IAM::Role',
+ Properties: {},
+ },
+ },
+ };
+
+ it('builds ResourceToImport list for online eval config', () => {
+ const configName = 'QualityMonitor';
+ const configId = 'oec-123';
+
+ const resourcesToImport: ResourceToImport[] = [];
+
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ configName
+ );
+
+ if (logicalId) {
+ resourcesToImport.push({
+ resourceType: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ logicalResourceId: logicalId,
+ resourceIdentifier: { OnlineEvaluationConfigId: configId },
+ });
+ }
+
+ expect(resourcesToImport).toHaveLength(1);
+ expect(resourcesToImport[0]!.resourceType).toBe('AWS::BedrockAgentCore::OnlineEvaluationConfig');
+ expect(resourcesToImport[0]!.logicalResourceId).toBe('OnlineEvalLogicalId');
+ expect(resourcesToImport[0]!.resourceIdentifier).toEqual({ OnlineEvaluationConfigId: 'oec-123' });
+ });
+
+ it('returns empty list when online eval config not found in template', () => {
+ const emptyTemplate: CfnTemplate = {
+ Resources: {
+ IAMRoleLogicalId: {
+ Type: 'AWS::IAM::Role',
+ Properties: {},
+ },
+ },
+ };
+
+ const logicalId = findLogicalIdByProperty(
+ emptyTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'QualityMonitor'
+ );
+
+ expect(logicalId).toBeUndefined();
+ });
+});
+
+// ============================================================================
+// buildImportTemplate Tests for Online Eval Config Resources
+// ============================================================================
+
+describe('buildImportTemplate with Online Eval Config', () => {
+ it('adds online eval config resource to deployed template with Retain deletion policy', () => {
+ const deployedTemplate: CfnTemplate = {
+ Resources: {
+ ExistingIAMRole: {
+ Type: 'AWS::IAM::Role',
+ Properties: { RoleName: 'ExistingRole' },
+ },
+ },
+ };
+
+ const synthTemplate: CfnTemplate = {
+ Resources: {
+ ExistingIAMRole: {
+ Type: 'AWS::IAM::Role',
+ Properties: { RoleName: 'ExistingRole' },
+ },
+ OnlineEvalLogicalId: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'QualityMonitor',
+ },
+ DependsOn: 'ExistingIAMRole',
+ },
+ },
+ };
+
+ const importTemplate = buildImportTemplate(deployedTemplate, synthTemplate, ['OnlineEvalLogicalId']);
+
+ // Verify online eval config resource was added
+ expect(importTemplate.Resources.OnlineEvalLogicalId).toBeDefined();
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.Type).toBe('AWS::BedrockAgentCore::OnlineEvaluationConfig');
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.DeletionPolicy).toBe('Retain');
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.UpdateReplacePolicy).toBe('Retain');
+
+ // DependsOn should be removed for import
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.DependsOn).toBeUndefined();
+
+ // Original resource should still be there
+ expect(importTemplate.Resources.ExistingIAMRole).toBeDefined();
+ });
+});

src/cli/commands/import/command.ts

@@ -1,14 +1,13 @@
 import { handleImport } from './actions';
+import { ANSI } from './constants';
+import { registerImportEvaluator } from './import-evaluator';
 import { registerImportMemory } from './import-memory';
+import { registerImportOnlineEval } from './import-online-eval';
 import { registerImportRuntime } from './import-runtime';
 import type { Command } from '@commander-js/extra-typings';
 import * as fs from 'node:fs';
 
-const green = '\x1b[32m';
-const yellow = '\x1b[33m';
-const cyan = '\x1b[36m';
-const dim = '\x1b[2m';
-const reset = '\x1b[0m';
+const { green, yellow, cyan, dim, reset } = ANSI;
 
 export const registerImport = (program: Command) => {
  const importCmd = program
@@ -151,4 +150,6 @@ export const registerImport = (program: Command) => {
  // Register subcommands for importing individual resource types from AWS
  registerImportRuntime(importCmd);
  registerImportMemory(importCmd);
+ registerImportEvaluator(importCmd);
+ registerImportOnlineEval(importCmd);
 };

src/cli/commands/import/__tests__/import-online-eval.test.ts

@@ -0,0 +1,368 @@
+/**
+ * Import Online Eval Config Unit Tests
+ *
+ * Covers:
+ * - extractAgentName: service name parsing
+ * - toOnlineEvalConfigSpec conversion: happy path, missing sampling, enableOnCreate
+ * - Template logical ID lookup for online eval configs
+ * - Phase 2 import resource list construction for online eval configs
+ */
+import type { GetOnlineEvalConfigResult } from '../../../aws/agentcore-control';
+import { extractAgentName, toOnlineEvalConfigSpec } from '../import-online-eval';
+import { buildImportTemplate, findLogicalIdByProperty, findLogicalIdsByType } from '../template-utils';
+import type { CfnTemplate } from '../template-utils';
+import type { ResourceToImport } from '../types';
+import { describe, expect, it } from 'vitest';
+
+// ============================================================================
+// extractAgentName Tests
+// ============================================================================
+
+describe('extractAgentName', () => {
+ it('extracts agent name from service name with .DEFAULT suffix', () => {
+ expect(extractAgentName(['my_agent.DEFAULT'])).toBe('my_agent');
+ });
+
+ it('extracts agent name with project prefix pattern', () => {
+ expect(extractAgentName(['testproject_my_agent.DEFAULT'])).toBe('testproject_my_agent');
+ });
+
+ it('returns full string when no dot suffix', () => {
+ expect(extractAgentName(['my_agent'])).toBe('my_agent');
+ });
+
+ it('returns undefined for empty array', () => {
+ expect(extractAgentName([])).toBeUndefined();
+ });
+
+ it('uses first service name when multiple provided', () => {
+ expect(extractAgentName(['agent_one.DEFAULT', 'agent_two.DEFAULT'])).toBe('agent_one');
+ });
+
+ it('handles service name with multiple dots', () => {
+ expect(extractAgentName(['my.agent.DEFAULT'])).toBe('my.agent');
+ });
+});
+
+// ============================================================================
+// toOnlineEvalConfigSpec Conversion Tests
+// ============================================================================
+
+describe('toOnlineEvalConfigSpec', () => {
+ it('maps online eval config with all fields', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-123',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-123',
+ configName: 'QualityMonitor',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ description: 'Monitor agent quality',
+ samplingPercentage: 50,
+ serviceNames: ['my_agent.DEFAULT'],
+ evaluatorIds: ['eval-456'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'QualityMonitor', 'my_agent', ['my_evaluator']);
+
+ expect(result.name).toBe('QualityMonitor');
+ expect(result.agent).toBe('my_agent');
+ expect(result.evaluators).toEqual(['my_evaluator']);
+ expect(result.samplingRate).toBe(50);
+ expect(result.description).toBe('Monitor agent quality');
+ expect(result.enableOnCreate).toBe(true);
+ });
+
+ it('omits enableOnCreate when execution status is DISABLED', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-456',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-456',
+ configName: 'DisabledConfig',
+ status: 'ACTIVE',
+ executionStatus: 'DISABLED',
+ samplingPercentage: 10,
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'DisabledConfig', 'agent', ['eval_one']);
+
+ expect(result.enableOnCreate).toBeUndefined();
+ });
+
+ it('omits description when not present', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-789',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-789',
+ configName: 'NoDesc',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ samplingPercentage: 25,
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'NoDesc', 'agent', ['eval_one']);
+
+ expect(result.description).toBeUndefined();
+ });
+
+ it('throws when sampling percentage is missing', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-no-sampling',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-no-sampling',
+ configName: 'NoSampling',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1'],
+ };
+
+ expect(() => toOnlineEvalConfigSpec(detail, 'NoSampling', 'agent', ['eval_one'])).toThrow(
+ 'has no sampling configuration'
+ );
+ });
+
+ it('supports multiple evaluator references', () => {
+ const detail: GetOnlineEvalConfigResult = {
+ configId: 'oec-multi',
+ configArn: 'arn:aws:bedrock-agentcore:us-west-2:123456789012:online-evaluation-config/oec-multi',
+ configName: 'MultiEval',
+ status: 'ACTIVE',
+ executionStatus: 'ENABLED',
+ samplingPercentage: 75,
+ serviceNames: ['agent.DEFAULT'],
+ evaluatorIds: ['eval-1', 'eval-2'],
+ };
+
+ const result = toOnlineEvalConfigSpec(detail, 'MultiEval', 'agent', [
+ 'local_eval',
+ 'arn:aws:bedrock-agentcore:us-west-2:123456789012:evaluator/eval-2',
+ ]);
+
+ expect(result.evaluators).toHaveLength(2);
+ expect(result.evaluators[0]).toBe('local_eval');
+ expect(result.evaluators[1]).toMatch(/^arn:/);
+ });
+});
+
+// ============================================================================
+// Template Logical ID Lookup Tests for Online Eval Configs
+// ============================================================================
+
+describe('Template Logical ID Lookup for Online Eval Configs', () => {
+ const synthTemplate: CfnTemplate = {
+ Resources: {
+ MyOnlineEvalConfig: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'QualityMonitor',
+ },
+ },
+ PrefixedOnlineEvalConfig: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'TestProject_PrefixedConfig',
+ },
+ },
+ MyAgentRuntime: {
+ Type: 'AWS::BedrockAgentCore::Runtime',
+ Properties: {
+ AgentRuntimeName: 'TestProject_my_agent',
+ },
+ },
+ MyIAMRole: {
+ Type: 'AWS::IAM::Role',
+ Properties: {
+ RoleName: 'MyRole',
+ },
+ },
+ },
+ };
+
+ it('finds online eval config logical ID by OnlineEvaluationConfigName property', () => {
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'QualityMonitor'
+ );
+ expect(logicalId).toBe('MyOnlineEvalConfig');
+ });
+
+ it('finds prefixed online eval config by full name', () => {
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'TestProject_PrefixedConfig'
+ );
+ expect(logicalId).toBe('PrefixedOnlineEvalConfig');
+ });
+
+ it('finds all online eval config logical IDs by type', () => {
+ const logicalIds = findLogicalIdsByType(synthTemplate, 'AWS::BedrockAgentCore::OnlineEvaluationConfig');
+ expect(logicalIds).toHaveLength(2);
+ expect(logicalIds).toContain('MyOnlineEvalConfig');
+ expect(logicalIds).toContain('PrefixedOnlineEvalConfig');
+ });
+
+ it('returns undefined for non-existent config name', () => {
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'nonexistent_config'
+ );
+ expect(logicalId).toBeUndefined();
+ });
+
+ it('falls back to single online eval config logical ID when name does not match', () => {
+ const singleConfigTemplate: CfnTemplate = {
+ Resources: {
+ OnlyConfig: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'some_config',
+ },
+ },
+ },
+ };
+
+ let logicalId = findLogicalIdByProperty(
+ singleConfigTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'different_name'
+ );
+
+ // Primary lookup fails
+ expect(logicalId).toBeUndefined();
+
+ // Fallback: if there's only one config resource, use it
+ if (!logicalId) {
+ const configLogicalIds = findLogicalIdsByType(
+ singleConfigTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig'
+ );
+ if (configLogicalIds.length === 1) {
+ logicalId = configLogicalIds[0];
+ }
+ }
+ expect(logicalId).toBe('OnlyConfig');
+ });
+});
+
+// ============================================================================
+// Phase 2 Resource Import List Construction for Online Eval Configs
+// ============================================================================
+
+describe('Phase 2: ResourceToImport List Construction for Online Eval Configs', () => {
+ const synthTemplate: CfnTemplate = {
+ Resources: {
+ OnlineEvalLogicalId: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'QualityMonitor',
+ },
+ },
+ IAMRoleLogicalId: {
+ Type: 'AWS::IAM::Role',
+ Properties: {},
+ },
+ },
+ };
+
+ it('builds ResourceToImport list for online eval config', () => {
+ const configName = 'QualityMonitor';
+ const configId = 'oec-123';
+
+ const resourcesToImport: ResourceToImport[] = [];
+
+ const logicalId = findLogicalIdByProperty(
+ synthTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ configName
+ );
+
+ if (logicalId) {
+ resourcesToImport.push({
+ resourceType: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ logicalResourceId: logicalId,
+ resourceIdentifier: { OnlineEvaluationConfigId: configId },
+ });
+ }
+
+ expect(resourcesToImport).toHaveLength(1);
+ expect(resourcesToImport[0]!.resourceType).toBe('AWS::BedrockAgentCore::OnlineEvaluationConfig');
+ expect(resourcesToImport[0]!.logicalResourceId).toBe('OnlineEvalLogicalId');
+ expect(resourcesToImport[0]!.resourceIdentifier).toEqual({ OnlineEvaluationConfigId: 'oec-123' });
+ });
+
+ it('returns empty list when online eval config not found in template', () => {
+ const emptyTemplate: CfnTemplate = {
+ Resources: {
+ IAMRoleLogicalId: {
+ Type: 'AWS::IAM::Role',
+ Properties: {},
+ },
+ },
+ };
+
+ const logicalId = findLogicalIdByProperty(
+ emptyTemplate,
+ 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ 'OnlineEvaluationConfigName',
+ 'QualityMonitor'
+ );
+
+ expect(logicalId).toBeUndefined();
+ });
+});
+
+// ============================================================================
+// buildImportTemplate Tests for Online Eval Config Resources
+// ============================================================================
+
+describe('buildImportTemplate with Online Eval Config', () => {
+ it('adds online eval config resource to deployed template with Retain deletion policy', () => {
+ const deployedTemplate: CfnTemplate = {
+ Resources: {
+ ExistingIAMRole: {
+ Type: 'AWS::IAM::Role',
+ Properties: { RoleName: 'ExistingRole' },
+ },
+ },
+ };
+
+ const synthTemplate: CfnTemplate = {
+ Resources: {
+ ExistingIAMRole: {
+ Type: 'AWS::IAM::Role',
+ Properties: { RoleName: 'ExistingRole' },
+ },
+ OnlineEvalLogicalId: {
+ Type: 'AWS::BedrockAgentCore::OnlineEvaluationConfig',
+ Properties: {
+ OnlineEvaluationConfigName: 'QualityMonitor',
+ },
+ DependsOn: 'ExistingIAMRole',
+ },
+ },
+ };
+
+ const importTemplate = buildImportTemplate(deployedTemplate, synthTemplate, ['OnlineEvalLogicalId']);
+
+ // Verify online eval config resource was added
+ expect(importTemplate.Resources.OnlineEvalLogicalId).toBeDefined();
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.Type).toBe('AWS::BedrockAgentCore::OnlineEvaluationConfig');
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.DeletionPolicy).toBe('Retain');
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.UpdateReplacePolicy).toBe('Retain');
+
+ // DependsOn should be removed for import
+ expect(importTemplate.Resources.OnlineEvalLogicalId!.DependsOn).toBeUndefined();
+
+ // Original resource should still be there
+ expect(importTemplate.Resources.ExistingIAMRole).toBeDefined();
+ });
+});

src/cli/commands/import/command.ts

@@ -1,14 +1,13 @@
 import { handleImport } from './actions';
+import { ANSI } from './constants';
+import { registerImportEvaluator } from './import-evaluator';
 import { registerImportMemory } from './import-memory';
+import { registerImportOnlineEval } from './import-online-eval';
 import { registerImportRuntime } from './import-runtime';
 import type { Command } from '@commander-js/extra-typings';
 import * as fs from 'node:fs';
 
-const green = '\x1b[32m';
-const yellow = '\x1b[33m';
-const cyan = '\x1b[36m';
-const dim = '\x1b[2m';
-const reset = '\x1b[0m';
+const { green, yellow, cyan, dim, reset } = ANSI;
 
 export const registerImport = (program: Command) => {
  const importCmd = program
@@ -151,4 +150,6 @@ export const registerImport = (program: Command) => {
  // Register subcommands for importing individual resource types from AWS
  registerImportRuntime(importCmd);
  registerImportMemory(importCmd);
+ registerImportEvaluator(importCmd);
+ registerImportOnlineEval(importCmd);
 };