MCP Business AI Transformation

from abc import ABC, abstractmethod from typing import Dict, Any, Optional, List import structlog from datetime import datetime logger = structlog.get_logger() class LLMResponse: """Response from LLM provider""" def __init__( self, content: str, model: str, tokens_used: int, cost: float, metadata: Dict[str, Any] = None ): self.content = content self.model = model self.tokens_used = tokens_used self.cost = cost self.metadata = metadata or {} self.timestamp = datetime.utcnow() class BaseLLMProvider(ABC): """Abstract base class for LLM providers""" def __init__(self, provider_name: str, config: Dict[str, Any] = None): self.provider_name = provider_name self.config = config or {} self.total_tokens_used = 0 self.total_cost = 0.0 @abstractmethod async def generate( self, prompt: str, max_tokens: int = 1000, temperature: float = 0.7, **kwargs ) -> LLMResponse: """Generate text using the LLM""" pass @abstractmethod async def generate_with_chat( self, messages: List[Dict[str, str]], max_tokens: int = 1000, temperature: float = 0.7, **kwargs ) -> LLMResponse: """Generate text using chat format""" pass @abstractmethod async def get_available_models(self) -> List[str]: """Get list of available models""" pass @abstractmethod async def estimate_cost(self, tokens: int, model: str) -> float: """Estimate cost for token usage""" pass def get_usage_stats(self) -> Dict[str, Any]: """Get usage statistics""" return { "provider": self.provider_name, "total_tokens_used": self.total_tokens_used, "total_cost": self.total_cost, "average_cost_per_token": self.total_cost / self.total_tokens_used if self.total_tokens_used > 0 else 0 } def _update_usage_stats(self, tokens_used: int, cost: float): """Update usage statistics""" self.total_tokens_used += tokens_used self.total_cost += cost logger.info( "LLM usage updated", provider=self.provider_name, tokens_used=tokens_used, cost=cost, total_tokens=self.total_tokens_used, total_cost=self.total_cost )