This tutorial focuses on how the researcher agent uses the tavily search tool, to get information of a specific thing from web and working along with the chart generator agent to create the code to get the graph out. Once that is sorted, the chart generator agent will send the graph back to the researcher to give out information, meanwhile once the graph is sorted, it will end the query by chart generator.

For this tutorial, we have used tavily, AzureDevOps, PythonREPL.

https://langchain-ai.github.io/langgraph/tutorials/multi_agent/multi-agent-collaboration/

In this code, since we weren’t using anthropic, we resorted to handling the envs in an env file and setting up the environment variables as below

import os 
from dotenv import load_dotenv 

load_dotenv()

os.environ['OPENAI_API_KEY'] = os.getenv("OPENAI_API_KEY")
os.environ['AZURE_OPENAI_API_KEY'] = os.getenv("AZURE_OPENAI_API_KEY")
os.environ['AZURE_OPENAI_ENDPOINT'] = os.getenv("AZURE_OPENAI_ENDPOINT")
os.environ['AZURE_OPENAI_API_VERSION'] = os.getenv("AZURE_OPENAI_API_VERSION")

#setting up langsmith
os.environ['LANGCHAIN_TRACING_V2'] = os.getenv("LANGCHAIN_TRACING_V2")
os.environ['LANGCHAIN_API_KEY'] = os.getenv("LANGCHAIN_API_KEY")
os.environ['LANGCHAIN_ENDPOINT'] = os.getenv("LANGCHAIN_ENDPOINT")
os.environ['LANGCHAIN_PROJECT'] = os.getenv("LANGCHAIN_PROJECT")

StateGraph of Researcher and ChartGenerator