@llm-dev-ops/agentics-cli 1.4.27 → 1.4.29

package/dist/bundled-agents/test-bench-agents/functions/index.ts

@@ -1,257 +0,0 @@
-/**
- * Cloud Function Entry Point - test-bench-agents
- *
- * Domain: test-bench
- * Function name: test-bench-agents
- * URL: https://us-central1-agentics-dev.cloudfunctions.net/test-bench-agents
- * Entry point: api
- *
- * Exposes 14 agents at /v1/test-bench/:agent routes.
- * Deployed via:
- *   gcloud functions deploy test-bench-agents \
- *     --runtime nodejs20 --trigger-http --allow-unauthenticated \
- *     --region us-central1 --project agentics-dev \
- *     --entry-point api --memory 512MB --timeout 120s
- */
-
-import express, { Request, Response, NextFunction } from 'express';
-import { randomUUID } from 'crypto';
-
-// Import all 14 agent handlers
-import { handler as benchmarkRunnerHandler } from '../agents/benchmark-runner/handler';
-import { handler as modelComparatorHandler } from '../agents/model-comparator/handler';
-import { handler as regressionDetectionHandler } from '../agents/regression-detection/handler';
-import { handler as qualityScoringHandler } from '../agents/quality-scoring/handler';
-import { handler as hallucinationDetectorHandler } from '../agents/hallucination-detector/handler';
-import { handler as faithfulnessVerificationHandler } from '../agents/faithfulness-verification/handler';
-import { handler as biasDetectionHandler } from '../agents/bias-detection/handler';
-import { handler as promptSensitivityHandler } from '../agents/prompt-sensitivity/handler';
-import { handler as outputConsistencyHandler } from '../agents/output-consistency/handler';
-import { handler as stressTestHandler } from '../agents/stress-test/handler';
-import { handler as redTeamHandler } from '../agents/red-team/handler';
-import { handler as adversarialPromptHandler } from '../agents/adversarial-prompt/handler';
-import { handler as goldenDatasetValidatorHandler } from '../agents/golden-dataset-validator/handler';
-import { handler as syntheticDataGeneratorHandler } from '../agents/synthetic-data-generator/handler';
-
-// =============================================================================
-// TYPES
-// =============================================================================
-
-interface EdgeFunctionRequest {
-  body: unknown;
-  headers: Record<string, string>;
-  method: string;
-  path: string;
-}
-
-interface EdgeFunctionResponse {
-  statusCode: number;
-  headers: Record<string, string>;
-  body: string;
-}
-
-type AgentHandler = (request: EdgeFunctionRequest) => Promise<EdgeFunctionResponse>;
-
-// =============================================================================
-// ROUTE MAPPING: slug -> { handler, layers }
-// =============================================================================
-
-const AGENT_ROUTES: Record<string, { handler: AgentHandler; name: string; layers: string[] }> = {
-  'benchmark': {
-    handler: benchmarkRunnerHandler,
-    name: 'Benchmark Runner',
-    layers: ['input_validation', 'provider_resolution', 'benchmark_execution', 'statistics_aggregation', 'scoring'],
-  },
-  'compare': {
-    handler: modelComparatorHandler,
-    name: 'Model Comparator',
-    layers: ['input_validation', 'model_selection', 'parallel_execution', 'comparison_analysis', 'ranking'],
-  },
-  'regression': {
-    handler: regressionDetectionHandler,
-    name: 'Regression Detection',
-    layers: ['input_validation', 'baseline_loading', 'regression_analysis', 'threshold_evaluation', 'scoring'],
-  },
-  'quality': {
-    handler: qualityScoringHandler,
-    name: 'Quality Scoring',
-    layers: ['input_validation', 'criteria_evaluation', 'quality_scoring', 'aggregation'],
-  },
-  'hallucination': {
-    handler: hallucinationDetectorHandler,
-    name: 'Hallucination Detection',
-    layers: ['input_validation', 'claim_extraction', 'fact_verification', 'hallucination_scoring'],
-  },
-  'faithfulness': {
-    handler: faithfulnessVerificationHandler,
-    name: 'Faithfulness Verification',
-    layers: ['input_validation', 'source_analysis', 'faithfulness_evaluation', 'scoring'],
-  },
-  'bias': {
-    handler: biasDetectionHandler,
-    name: 'Bias Detection',
-    layers: ['input_validation', 'demographic_analysis', 'bias_detection', 'scoring'],
-  },
-  'prompt-sensitivity': {
-    handler: promptSensitivityHandler,
-    name: 'Prompt Sensitivity',
-    layers: ['input_validation', 'prompt_variation', 'sensitivity_measurement', 'scoring'],
-  },
-  'consistency': {
-    handler: outputConsistencyHandler,
-    name: 'Output Consistency',
-    layers: ['input_validation', 'repeated_execution', 'consistency_analysis', 'scoring'],
-  },
-  'stress': {
-    handler: stressTestHandler,
-    name: 'Stress Test',
-    layers: ['input_validation', 'scenario_execution', 'failure_analysis', 'robustness_scoring'],
-  },
-  'red-team': {
-    handler: redTeamHandler,
-    name: 'Red Team',
-    layers: ['input_validation', 'attack_generation', 'attack_execution', 'resistance_evaluation', 'safety_scoring'],
-  },
-  'adversarial': {
-    handler: adversarialPromptHandler,
-    name: 'Adversarial Prompt',
-    layers: ['input_validation', 'prompt_generation', 'categorization', 'severity_ranking'],
-  },
-  'golden-dataset': {
-    handler: goldenDatasetValidatorHandler,
-    name: 'Golden Dataset Validator',
-    layers: ['input_validation', 'dataset_loading', 'validation_execution', 'accuracy_scoring'],
-  },
-  'synthetic-data': {
-    handler: syntheticDataGeneratorHandler,
-    name: 'Synthetic Data Generator',
-    layers: ['input_validation', 'schema_analysis', 'data_generation', 'quality_validation'],
-  },
-};
-
-// =============================================================================
-// EXPRESS APP
-// =============================================================================
-
-const app = express();
-
-// JSON body parser
-app.use(express.json({ limit: '10mb' }));
-
-// CORS middleware
-app.use((_req: Request, res: Response, next: NextFunction) => {
-  res.setHeader('Access-Control-Allow-Origin', '*');
-  res.setHeader('Access-Control-Allow-Methods', 'GET, POST, OPTIONS');
-  res.setHeader(
-    'Access-Control-Allow-Headers',
-    'X-Correlation-ID, X-API-Version, Content-Type, Authorization'
-  );
-  if (_req.method === 'OPTIONS') {
-    res.status(204).end();
-    return;
-  }
-  next();
-});
-
-// =============================================================================
-// HEALTH ENDPOINT
-// =============================================================================
-
-app.get('/health', (_req: Request, res: Response) => {
-  res.json({
-    healthy: true,
-    service: 'test-bench-agents',
-    agents: 14,
-  });
-});
-
-// =============================================================================
-// AGENT ROUTES: POST /v1/test-bench/:agent
-// =============================================================================
-
-app.post('/v1/test-bench/:agent', async (req: Request, res: Response) => {
-  const agentSlug = req.params.agent;
-  const route = AGENT_ROUTES[agentSlug];
-  const traceId = randomUUID();
-
-  if (!route) {
-    res.status(404).json({
-      error: 'Agent not found',
-      agent: agentSlug,
-      available_agents: Object.keys(AGENT_ROUTES),
-      execution_metadata: {
-        trace_id: traceId,
-        timestamp: new Date().toISOString(),
-        service: 'test-bench-agents',
-      },
-      layers_executed: ['input_validation'],
-    });
-    return;
-  }
-
-  try {
-    // Adapt Express request to EdgeFunctionRequest
-    const edgeRequest: EdgeFunctionRequest = {
-      body: req.body,
-      headers: Object.fromEntries(
-        Object.entries(req.headers)
-          .filter((entry): entry is [string, string] => typeof entry[1] === 'string')
-      ),
-      method: 'POST',
-      path: req.path,
-    };
-
-    // Call agent handler
-    const edgeResponse = await route.handler(edgeRequest);
-
-    // Parse response body to inject required metadata
-    let responseBody: Record<string, unknown>;
-    try {
-      responseBody = JSON.parse(edgeResponse.body);
-    } catch {
-      responseBody = { raw: edgeResponse.body };
-    }
-
-    // Inject execution_metadata (required by CLI contract validator)
-    responseBody.execution_metadata = {
-      trace_id: traceId,
-      timestamp: new Date().toISOString(),
-      service: 'test-bench-agents',
-    };
-
-    // Inject layers_executed (required by CLI validator)
-    responseBody.layers_executed = route.layers;
-
-    // Forward agent headers
-    for (const [key, value] of Object.entries(edgeResponse.headers)) {
-      if (key.toLowerCase() !== 'content-type') {
-        res.setHeader(key, value);
-      }
-    }
-
-    res.status(edgeResponse.statusCode).json(responseBody);
-
-  } catch (err) {
-    const error = err instanceof Error ? err : new Error(String(err));
-    res.status(500).json({
-      success: false,
-      error: {
-        code: 'EXECUTION_ERROR',
-        message: error.message,
-        recoverable: false,
-      },
-      execution_metadata: {
-        trace_id: traceId,
-        timestamp: new Date().toISOString(),
-        service: 'test-bench-agents',
-      },
-      layers_executed: ['input_validation', 'error_handling'],
-    });
-  }
-});
-
-// =============================================================================
-// EXPORT: Cloud Function entry point
-// =============================================================================
-
-export const api = app;