UP Coding Agent

UP Coding Agent - An intelligent TypeScript-based coding agent with LLM integration support.

Created by Martin Yeung

🚀 Features

Core Features

• Code Analysis: Automatic symbol extraction, complexity calculation, multi-language support
• Intelligent Completion: Context-aware variable/function completion, syntax suggestions
• Code Refactoring: Quick fixes, optimization suggestions, import organization
• Batch Processing: Multi-file analysis, cross-file symbol lookup

LLM Integration (NEW!)

• Multi-Provider Support: OpenAI, Anthropic, and more
• Code Generation: Generate secure, high-performance code using LLMs
• Security Analysis: Advanced vulnerability detection using AI
• Performance Optimization: Intelligent code optimization suggestions
• Flexible Configuration: Use LLM when needed, fallback to traditional methods

📦 Installation

npm install up-coding-agent

🎯 Quick Start

Basic Usage (Without LLM)

import { CodingAgent } from 'up-coding-agent';

const agent = new CodingAgent();

// Analyze code
const analysis = await agent.analyzeFile(code, 'file.ts');
console.log(analysis.symbols);

// Get completions
const completions = await agent.getCompletions(code, position, 'file.ts');

Advanced Usage (With LLM)

import { CodingAgent } from 'up-coding-agent';

const agent = new CodingAgent({ enableLLM: true });

// Configure LLM provider
agent.registerLLMProvider('openai', {
  provider: 'openai',
  apiKey: 'your-api-key',
  modelName: 'gpt-4'
});

// Generate code using LLM
const generatedCode = await agent.generateCodeWithLLM({
  language: 'typescript',
  prompt: 'Create a secure authentication service',
  constraints: ['Use bcrypt', 'Implement rate limiting']
});

// Security analysis using LLM
const securityResult = await agent.analyzeSecurityWithLLM({
  code: yourCode,
  language: 'javascript',
  checkTypes: ['injection', 'auth', 'all']
});

🔧 LLM Provider Setup

Environment Variables

Set your API keys in environment variables:

# OpenAI
export OPENAI_API_KEY="sk-your-openai-key"
export OPENAI_MODEL="gpt-4"  # optional

# Anthropic
export ANTHROPIC_API_KEY="sk-your-anthropic-key"
export ANTHROPIC_MODEL="claude-3-opus-20240229"  # optional

Supported Providers

• OpenAI: GPT-4, GPT-3.5 Turbo
• Anthropic: Claude 3 Opus, Sonnet, Haiku
• Custom: Any compatible API endpoint

📚 API Reference

CodingAgent Class

Constructor Options

interface AgentConfig {
  maxContextLength: number;      // Maximum context length
  supportLanguages: string[];    // Supported languages
  enableAutoImport: boolean;     // Auto-import suggestions
  enableCodeGeneration: boolean; // Code generation features
  enableLLM: boolean;            // Enable LLM features (default: true)
  temperature: number;           // Temperature for generation (0-1)
}

LLM Methods

// Check if LLM is available
agent.isLLMAvailable(): boolean

// Register LLM provider
agent.registerLLMProvider(name: string, config: LLMConfig): void

// Generate code with LLM
agent.generateCodeWithLLM(request: CodeGenerationRequest): Promise<string>

// Security analysis with LLM
agent.analyzeSecurityWithLLM(request: SecurityAnalysisRequest): Promise<SecurityAnalysisResult>

// Performance optimization with LLM
agent.optimizePerformanceWithLLM(request: PerformanceOptimizationRequest): Promise<PerformanceOptimizationResult>

// Chat with LLM
agent.chatWithLLM(messages: Array, options?: LLMOptions): Promise<any>

🧪 Testing

npm test

📚 Examples

Run the included examples:

# Basic example
npm run example

# Advanced features
npm run advanced

# Real-world scenarios
npm run realworld

# LLM integration demo
npm run llm-example

🌟 Supported Languages

• TypeScript/JavaScript (TS, TSX, JS, JSX)
• Python (PY)
• Java (JAVA)
• C/C++ (CPP, CC, CXX, C, H)
• C# (CS)

🔧 Configuration

Traditional Mode (No LLM)

const agent = new CodingAgent({
  enableLLM: false,  // Disable LLM features
  maxContextLength: 4096
});

LLM Mode

const agent = new CodingAgent({
  enableLLM: true,   // Enable LLM features
  temperature: 0.7
});

// Configure providers
agent.registerLLMProvider('openai', {
  provider: 'openai',
  apiKey: process.env.OPENAI_API_KEY,
  modelName: 'gpt-4'
});

🛠️ Development

Setup Development Environment

git clone https://github.com/martinyeung/up-coding-agent.git
cd up-coding-agent
npm install

Build

npm run build

Publish

npm publish

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

1. Fork the project
2. Create your feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add some AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

• Inspired by Cursor and other intelligent programming assistants
• Built with TypeScript for type safety and better developer experience
• Designed to be lightweight and extensible
• LLM integration allows for even more powerful code analysis and generation

📞 Support

For issues and questions, please visit our GitHub Issues.

Made by Martin Yeung

UP Coding Agent - Your intelligent coding companion with LLM-powered features!