Define LangGraph State and Initial Agent Nodes

Using LangGraph, define the `AgentState` for a content strategy workflow, including messages, market data, and content drafts. Create initial nodes for a 'MarketAnalyzer' agent and a 'ContentGenerator' agent. The MarketAnalyzer will use OpenAI o4-mini to process incoming market trends, and the ContentGenerator will use Writer for content drafting. Implement a basic graph where the MarketAnalyzer passes its findings to the ContentGenerator. ```python
from typing import TypedDict, Annotated, List
from langgraph.graph import StateGraph, END
from langchain_core.messages import BaseMessage, HumanMessage, AIMessage class AgentState(TypedDict): messages: Annotated[List[BaseMessage], lambda x, y: x + y] # Accumulate messages market_data: str content_draft: str def market_analyzer_node(state: AgentState) -> AgentState: # Use OpenAI o4-mini to analyze state['market_data'] # Simulate analysis for now analysis_result = f"Analysis of market data: {state['market_data']}" return {'messages': [AIMessage(content=analysis_result)], 'market_data': analysis_result} def content_generator_node(state: AgentState) -> AgentState: # Use Writer to generate content based on state['market_data'] # Simulate content generation for now generated_content = f"Content generated based on: {state['market_data']}" return {'messages': [AIMessage(content=generated_content)], 'content_draft': generated_content} workflow = StateGraph(AgentState)
workflow.add_node("market_analyzer", market_analyzer_node)
workflow.add_node("content_generator", content_generator_node) # Your task: Add edges to connect these nodes and define the entry point.
```