Enhanced MyCoder v2.2.0

Multi-API AI Development Assistant with Q9550 Thermal Management

Enhanced MyCoder v2.2.0 is a comprehensive AI development assistant featuring modular multi-provider architecture, Q9550 thermal management, agent orchestration, and FEI-inspired architecture. Built for production environments requiring high availability and thermal safety.

🇨🇿 Česká verze (Czech Version)

🚀 Quick Start

Installation

git clone https://github.com/milhy545/MyCoder-v2.0.git
cd MyCoder-v2.0
pip install -r requirements.txt
# Or using Poetry (Recommended)
poetry install

Basic Usage

from mycoder import EnhancedMyCoderV2
from pathlib import Path

# Basic configuration
config = {
    "claude_oauth": {"enabled": True},
    "ollama_local": {"enabled": True},
    "thermal": {"enabled": True, "max_temp": 75}
}

# Initialize MyCoder
mycoder = EnhancedMyCoderV2(
    working_directory=Path("."),
    config=config
)

# Start processing
await mycoder.initialize()

response = await mycoder.process_request(
    "Analyze this Python file and suggest optimizations",
    files=[Path("example.py")]
)

print(f"Response: {response['content']}")
print(f"Provider: {response['provider']}")
print(f"Cost: ${response['cost']}")

Interactive CLI

poetry run mycoder

Commands:

• /setup - Configure providers and keys
• /providers - List available providers
• /plan <task> - Generate implementation plan
• /voice start - Start dictation mode

🏗️ Architecture

Modular Provider System

MyCoder now supports a wide range of AI providers through a modular interface:

LLM Providers:

• Claude Anthropic API (Primary, High Quality)
• Claude OAuth (Authenticated CLI)
• Google Gemini (High Speed, Large Context)
• AWS Bedrock (Enterprise, Claude/Titan)
• OpenAI (GPT-4o, o1)
• X.AI (Grok)
• Mistral AI (Open/Commercial)
• HuggingFace (Inference API)
• Ollama (Local/Remote/Termux)
• Mercury (Inception Labs)

TTS Providers (Text-to-Speech):

• Azure Speech (High Quality Neural)
• Amazon Polly (Neural/Standard)
• ElevenLabs (Premium Cloning)
• gTTS (Google Translate)
• Local (pyttsx3, espeak)

STT Providers (Speech-to-Text):

• Whisper (OpenAI API / Local)
• Google Gemini (Multimodal)
• Azure Speech (Real-time)

FEI-Inspired Components

• Tool Registry Pattern: Centralized tool management with execution contexts
• Service Layer Pattern: Clean separation between API providers and business logic
• Event-Based Architecture: Reactive system with health monitoring and thermal awareness

Q9550 Thermal Management

Integrated thermal monitoring and throttling for Intel Q9550 processors:

• Temperature Monitoring: Real-time CPU temperature tracking
• Automatic Throttling: Reduces AI workload when temperature exceeds 75°C
• Emergency Protection: Hard shutdown at 85°C to prevent hardware damage
• PowerManagement Integration: Uses existing Q9550 thermal scripts

🔧 Configuration

Environment Variables

# API Keys
export ANTHROPIC_API_KEY="sk-..."
export GEMINI_API_KEY="AIza..."
export OPENAI_API_KEY="sk-..."
export XAI_API_KEY="xai-..."
export MISTRAL_API_KEY="..."
export HF_TOKEN="hf_..."
export ELEVENLABS_API_KEY="..."
export AZURE_SPEECH_KEY="..."
export AZURE_SPEECH_REGION="eastus"

# AWS Credentials (if using Bedrock/Polly)
export AWS_ACCESS_KEY_ID="..."
export AWS_SECRET_ACCESS_KEY="..."
export AWS_REGION="us-east-1"

# System Configuration  
export MYCODER_DEBUG=1
export MYCODER_THERMAL_MAX_TEMP=75
export MYCODER_PREFERRED_PROVIDER=claude_oauth

Configuration File

Create mycoder_config.json:

{
  "claude_anthropic": {
    "enabled": true,
    "timeout_seconds": 30,
    "model": "claude-3-5-sonnet-20241022"
  },
  "gemini": {
    "enabled": true,
    "rate_limit_rpm": 15,
    "rate_limit_rpd": 1500
  },
  "aws_bedrock": {
    "enabled": true,
    "model": "anthropic.claude-3-sonnet-20240229-v1:0"
  },
  "text_to_speech": {
    "enabled": true,
    "provider": "azure",
    "voice": "en-US-JennyNeural"
  }
}

🛠️ Features

Multi-API Provider Support

• Intelligent Fallback: Automatic failover between providers
• Health Monitoring: Real-time provider status tracking
• Cost Optimization: Prefer free/cheaper providers when available
• Performance Metrics: Track response times and success rates
• Circuit Breaker & Rate Limiting: Resilient API management with persistent rate limiting (RPM/RPD)

Thermal Management (Q9550)

• Hardware Integration: Direct integration with Q9550 thermal sensors
• Proactive Throttling: Prevent thermal damage before it occurs
• Performance Scaling: Adjust AI workload based on temperature
• System Protection: Emergency shutdown for critical temperatures

Tool Registry System

• Modular Tools: File operations, MCP integration, thermal monitoring
• Execution Contexts: Secure sandboxed tool execution
• Permission System: Role-based access control for tools
• Performance Monitoring: Track tool usage and performance

Session Management

• Persistent Sessions: Maintain conversation context across requests
• Provider Transitions: Seamless switching between API providers
• Automatic Cleanup: Memory-efficient session management
• Recovery Support: Restore sessions after system restart

📊 Testing

Comprehensive Test Suite

• Unit Tests (90% pass rate): Core component functionality
• Integration Tests (90% pass rate): Real-world scenarios
• Functional Tests (95% pass rate): End-to-end workflows
• Stress Tests (80% pass rate): System limits and edge cases

Running Tests

# All tests
python -m pytest tests/ -v

# Specific test types
python -m pytest tests/unit/ -v        # Unit tests
python -m pytest tests/integration/ -v # Integration tests
python -m pytest tests/functional/ -v  # Functional tests

# Stress testing
python tests/stress/run_stress_tests.py --all
python tests/stress/run_stress_tests.py --thermal  # Q9550 required

# Interactive testing
python tests/functional/test_mycoder_live.py --interactive

📄 License

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

🙏 Acknowledgments

• Anthropic, Google, OpenAI, AWS, Microsoft, ElevenLabs for AI services
• Ollama for local LLM infrastructure
• Intel Q9550 community for thermal management insights
• FEI architectural patterns inspiration

---

Made with ❤️ for the AI development community

Enhanced MyCoder v2.2.0 - Where AI meets thermal responsibility