Complete single-stage optimization implementation with session reuse

README.md

@@ -81,6 +81,7 @@ Options:
   -v, --version        output the version number
   -p, --port <port>    port to run server on (default: 8000)
   -d, --debug          enable debug mode (runs in foreground)
+  -m, --mock           enable mock mode (simulates Claude CLI responses)
   -k, --api-key <key>  set API key for endpoint protection
   -n, --no-interactive disable interactive API key setup
   -P, --production     enable production server management features
@@ -154,6 +155,16 @@ wrapper --stop
 wrapper -s                         # shorthand
 ```
 
+## 🎭 Mock Mode
+
+Develop and test without Claude CLI dependency! Mock mode provides ultra-fast simulation (300x faster) with complete API compatibility.
+
+```bash
+wrapper --mock    # Enable mock mode
+```
+
+📖 **[Mock Mode Guide](docs/MOCK_MODE.md)** - Complete documentation and configuration.
+
 ## 📚 Documentation
 
 📖 **[Full Documentation](docs/README.md)** - Comprehensive guide with detailed examples, production deployment, troubleshooting, and advanced configuration.

app/README.md

@@ -84,6 +84,7 @@ Options:
   -H, --health-monitoring enable health monitoring system
   -s, --stop           stop background server
   -t, --status         check background server status
+  -m, --mock           enable mock mode for testing (instant responses)
   -h, --help           display help for command
 ```
 
@@ -101,6 +102,34 @@ wrapper --api-key my-secure-key
 wrapper -k my-secure-key           # shorthand
 ```
 
+### Mock Mode for Testing
+
+**Mock mode provides instant responses for performance testing and development:**
+
+```bash
+# Enable mock mode for testing
+wrapper --mock
+wrapper -m                         # shorthand
+
+# Combine with other options
+wrapper --mock --debug             # mock mode with debug output
+wrapper --mock --port 9999         # mock mode on custom port
+```
+
+Mock mode returns realistic Claude CLI responses instantly without making actual API calls, perfect for:
+- Performance testing and load testing
+- Development and debugging
+- CI/CD pipeline testing
+- Offline development environments
+
+**Mock Mode Features:**
+- Instant response generation (no API delays)
+- Realistic Claude CLI JSON response format
+- Streaming support with word-by-word content deltas
+- Automatic token calculation based on prompt length
+- Unique session ID generation for each request
+- Compatible with all existing endpoints and authentication
+
 ## 📡 API Endpoints
 
 | Method   | Endpoint                    | Description                                   |

app/jest.config.js

@@ -1,11 +1,4 @@
 module.exports = {
-  // Custom reporter with automatic log cleanup and organized results
-  reporters: [
-    "default",
-    ["<rootDir>/tests/scripts/custom-reporter.js", {}],
-    ["<rootDir>/tests/scripts/verbose-reporter.js", {}]
-  ],
-
   // Projects setup for organized test types
   projects: [
     "<rootDir>/tests/jest.unit.config.js",

app/package.json

@@ -24,6 +24,8 @@
     "build": "tsc > build.log 2>&1",
     "prepublishOnly": "npm run build",
     "start": "node dist/cli.js",
+    "stop": "node dist/cli.js --stop",
+    "status": "node dist/cli.js --status",
     "dev": "ts-node src/cli.ts",
     "test": "jest",
     "test:unit": "jest --config tests/jest.unit.config.js",

app/src/api/middleware/model-validation.ts

@@ -2,7 +2,7 @@ import { Request, Response, NextFunction } from 'express';
 import { logger } from '../../utils/logger';
 
 // Valid Claude models - only confirmed models from CLI reference
-const VALID_CLAUDE_MODELS = ['sonnet', 'opus'];
+const VALID_CLAUDE_MODELS = ['sonnet', 'opus', 'haiku'];
 
 export class ModelValidationError extends Error {
   public readonly statusCode: number;

app/src/api/routes/chat.ts

@@ -1,5 +1,5 @@
 import { Router, Request, Response } from 'express';
-import { CoreWrapper } from '../../core/wrapper';
+import { sharedCoreWrapper } from '../../core/shared-wrapper';
 import { StreamingHandler } from '../../streaming/handler';
 import { OpenAIRequest } from '../../types';
 import { InvalidRequestError } from '../../utils/errors';
@@ -9,7 +9,6 @@ import { modelValidationMiddleware } from '../middleware/model-validation';
 import { logger } from '../../utils/logger';
 
 const router = Router();
-const coreWrapper = new CoreWrapper();
 const streamingHandler = new StreamingHandler();
 
 // Apply validation and streaming middleware to chat completions
@@ -61,7 +60,7 @@ router.post('/v1/chat/completions',
       });
 
       // Handle non-streaming request
-      const response = await coreWrapper.handleChatCompletion(request);
+      const response = await sharedCoreWrapper.handleChatCompletion(request);
 
       logger.info('Chat completion request completed successfully', {
         requestId: response.id,

app/src/api/routes/health.ts

@@ -1,5 +1,6 @@
 import { Router, Request, Response } from 'express';
 import { asyncHandler } from '../middleware/error';
+import { EnvironmentManager } from '../../config/env';
 import * as packageJson from '../../../package.json';
 
 const router = Router();
@@ -10,7 +11,8 @@ router.get('/health', asyncHandler(async (_req: Request, res: Response) => {
     service: packageJson.name,
     version: packageJson.version,
     description: packageJson.description,
-    timestamp: new Date().toISOString()
+    timestamp: new Date().toISOString(),
+    mock_mode: EnvironmentManager.isMockMode()
   });
 }));