Posm/structured-planning-ai

Implementing a Structured Planning AI Agent with LlamaIndex

1. Set up the environment

   • Skip this step if you have already set up the environment

   python -m venv .venv
   source .venv/bin/activate
   

2. Setup LlamaIndex

   pip install llama-index
   

3. Create a python file

   touch worker.py
   

   Or

   echo. > worker.py
   

4. Open the file in VSCode

   code worker.py
   

5. Add the needed imports

   from llama_index.core.tools import FunctionTool
   from llama_index.llms.openai import OpenAI
   from llama_index.core.agent import (
       StructuredPlannerAgent,
       FunctionCallingAgentWorker,
   )
   

6. Define the function

   def multiply(a: int, b: int) -> int:
    """Multiply two integers and returns the result integer"""
    return a * b
   

7. Define and configure the worker agent

   multiply_tool = FunctionTool.from_defaults(fn=multiply)
   llm = OpenAI(model="gpt-4o-mini")
   worker = FunctionCallingAgentWorker.from_tools([multiply_tool], llm=llm, verbose=True)
   worker_agent = StructuredPlannerAgent(worker, [multiply_tool], verbose=True)
   

8. Test the worker agent

   worker_agent.chat("Solve the equation x = 123 * (x + 2y + 3)")
   

9. Create .env file & add api key

   OPENAI_API_KEY="<your_api_key>"
   

10. Run the agent

python worker.py