🧾 Prompt

Build a multi-agent AI application in Python where different agents collaborate to solve a task.

Each agent should have a specific role, such as planning, researching, coding, summarizing, or reviewing.

The system should accept a user request, delegate subtasks to the appropriate agents, and combine their outputs into a final response.

🛠 Herramientas

• Python
• LLM API or local model
• LangChain / CrewAI / Autogen / similar framework
• FastAPI (optional)
• dotenv
• Tools layer (web search mock, calculator, file reader, etc.)

💡 Pistas

🧠 Idea base

No pienses en “muchos chats”, piensa en:

• 1 coordinator agent
• several specialist agents
• shared context

🔧 Roles posibles

• Planner Agent → divides the task
• Research Agent → gathers information
• Writer Agent → drafts the answer
• Reviewer Agent → checks quality