Una implementación de código para construir un sistema de investigación de múltiples agentes con agentes de OpenAI, herramientas de funciones, transferencias y memoria de sesión

!pip install openai-agents python-dotenv


import asyncio
import json
from datetime import datetime
from agents import Agent, Runner, function_tool, SQLiteSession
import os


os.environ['OPENAI_API_KEY'] = 'Use Your Own API Key'

@function_tool
def web_search(query: str, max_results: int = 3) -> str:
   """Simulate web search results for demonstration"""
   results = [
       f"Result 1 for '{query}': Latest findings show significant developments...",
       f"Result 2 for '{query}': Research indicates new approaches in this field...",
       f"Result 3 for '{query}': Expert analysis suggests important implications..."
   ]
   return f"Search results for '{query}':\n" + "\n".join(results[:max_results])


@function_tool
def analyze_data(data: str, analysis_type: str = "summary") -> str:
   """Analyze provided data with different analysis types"""
   analyses = {
       "summary": f"Summary: The data contains {len(data.split())} key points with main themes around innovation and efficiency.",
       "detailed": f"Detailed Analysis: Breaking down the {len(data)} characters of data reveals patterns in methodology and conclusions.",
       "trends": f"Trend Analysis: Current data suggests upward trajectory with 3 major inflection points identified."
   }
   return analyses.get(analysis_type, "Analysis complete: Standard evaluation performed.")


@function_tool
def save_research(title: str, content: str, category: str = "general") -> str:
   """Save research findings to a structured format"""
   timestamp = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
   research_entry = {
       "title": title,
       "content": content,
       "category": category,
       "timestamp": timestamp,
       "id": f"research_{len(content) % 1000}"
   }
   return f"✅ Research saved: '{title}' in category '{category}' at {timestamp}"

research_agent = Agent(
   name="Research Specialist",
   instructions="""You are an expert researcher who:
   - Conducts thorough web searches on any topic
   - Analyzes information critically and objectively
   - Identifies key insights and patterns
   - Always uses tools to gather and analyze data before responding""",
   tools=[web_search, analyze_data]
)


analyst_agent = Agent(
   name="Data Analyst",
   instructions="""You are a senior data analyst who:
   - Takes research findings and performs deep analysis
   - Identifies trends, patterns, and actionable insights
   - Creates structured summaries and recommendations
   - Uses analysis tools to enhance understanding""",
   tools=[analyze_data, save_research]
)


coordinator_agent = Agent(
   name="Research Coordinator",
   instructions="""You are a research coordinator who:
   - Manages multi-step research projects
   - Delegates tasks to appropriate specialists
   - Synthesizes findings from multiple sources
   - Makes final decisions on research direction
   - Handoff to research_agent for initial data gathering
   - Handoff to analyst_agent for detailed analysis""",
   handoffs=[research_agent, analyst_agent],
   tools=[save_research]
)

async def run_advanced_research_workflow():
   """Demonstrates a complete multi-agent research workflow"""
  
   session = SQLiteSession("research_session_001")
  
   print("🚀 Starting Advanced Multi-Agent Research System")
   print("=" * 60)
  
   research_topic = "artificial intelligence in healthcare 2024"
  
   print(f"\n📋 PHASE 1: Initiating research on '{research_topic}'")
   result1 = await Runner.run(
       coordinator_agent,
       f"I need comprehensive research on '{research_topic}'. Please coordinate a full research workflow including data gathering, analysis, and final report generation.",
       session=session
   )
   print(f"Coordinator Response: {result1.final_output}")
  
   print(f"\n📊 PHASE 2: Requesting detailed trend analysis")
   result2 = await Runner.run(
       coordinator_agent,
       "Based on the previous research, I need a detailed trend analysis focusing on emerging opportunities and potential challenges. Save the final analysis for future reference.",
       session=session
   )
   print(f"Analysis Response: {result2.final_output}")
  
   print(f"\n🔬 PHASE 3: Direct specialist analysis")
   result3 = await Runner.run(
       analyst_agent,
       "Perform a detailed analysis of the healthcare AI market, focusing on regulatory challenges and market opportunities. Categorize this as 'market_analysis'.",
       session=session
   )
   print(f"Specialist Response: {result3.final_output}")
  
   print("\n✅ Research workflow completed successfully!")
   return result1, result2, result3


async def run_focused_analysis():
   """Shows focused single-agent capabilities"""
  
   print("\n🎯 FOCUSED ANALYSIS DEMO")
   print("-" * 40)
  
   result = await Runner.run(
       research_agent,
       "Research in quantum computing and analyze the key breakthroughs from 2024.",
       max_turns=5
   )
  
   print(f"Focused Analysis Result: {result.final_output}")
   return result


def quick_research_sync(topic: str):
   """Synchronous research for quick queries"""
  
   print(f"\n⚡ QUICK SYNC RESEARCH: {topic}")
   print("-" * 40)
  
   result = Runner.run_sync(
       research_agent,
       f"Quickly research {topic} and provide 3 key insights."
   )
  
   print(f"Quick Result: {result.final_output}")
   return result

async def main():
   """Main function demonstrating all capabilities"""
  
   print("🤖 OpenAI Agents SDK - Advanced Tutorial")
   print("Building a Multi-Agent Research System")
   print("=" * 60)
  
   try:
       await run_advanced_research_workflow()
      
       await run_focused_analysis()
      
       quick_research_sync("blockchain adoption in enterprise")
      
       print("\n🎉 Tutorial completed successfully!")
       print("\nKey Features Demonstrated:")
       print("✅ Multi-agent coordination with handoffs")
       print("✅ Custom function tools")
       print("✅ Session memory for conversation continuity")
       print("✅ Async and sync execution patterns")
       print("✅ Structured workflows with max_turns control")
       print("✅ Specialized agent roles and capabilities")
      
   except Exception as e:
       print(f"❌ Error: {e}")
       print("\nTroubleshooting tips:")
       print("- Ensure OPENAI_API_KEY is set correctly")
       print("- Check internet connection")
       print("- Verify openai-agents package is installed")


if __name__ == "__main__":
   import nest_asyncio
   nest_asyncio.apply()
  
   asyncio.run(main())


def create_custom_agent(name: str, role: str, tools_list: list = None):
   """Helper function to create custom agents quickly"""
   return Agent(
       name=name,
       instructions=f"You are a {role} who provides expert assistance.",
       tools=tools_list or []
   )


custom_agent = create_custom_agent("Code Reviewer", "senior software engineer", [analyze_data])
result = Runner.run_sync(custom_agent, "Review this Python code for best practices")


print("\n📚 Tutorial Notes:")
print("- Modify research topics and agent instructions to explore different use cases")
print("- Add your own custom tools using the @function_tool decorator")
print("- Experiment with different agent handoff patterns")
print("- Use sessions for multi-turn conversations")
print("- Perfect for Colab - just add your OpenAI API key and run!")