Conversation with a ReAct agent

Designing Agentic Systems with LangChain

Dilini K. Sumanapala, PhD

Founder & AI Engineer, Genverv, Ltd.

Conversation

The area of a rectangle with sides 5 and 7 is 35 square units.

  • Validating answers

  • User: "What is the area of a rectangle with sides 5 and 7?"

  • Agent: "The area of a rectangle with sides 5 and 7 is 35 square units."
Designing Agentic Systems with LangChain

Conversation

tools = [rectangle_area]
query = "What is the area of a rectangle with sides 14 and 4?"


# Create the ReAct agent
app = create_react_agent(model, tools)


# Invoke the agent with a query and store the messages
response = app.invoke({"messages": [("human", query)]})


# Define and print the input and output messages
print({
    "user_input": query,
    "agent_output": response["messages"][-1].content})
Designing Agentic Systems with LangChain

Conversation output

{'user_input': 'What is the area of a rectangle with sides 14 and 4?',
 'agent_output': 'The area of a rectangle with sides 14 and 4 is 56 
 square units.'}
Designing Agentic Systems with LangChain

Follow-up questions

  • Follow-up:
    • User: "What about one with sides 12 and 14?"
  • Conversation history:
    • User: "What is the area of a rectangle with sides 5 and 7?"
    • Agent: "The area of a rectangle with sides 5 and 7 is 35 square units."
    • User: "What about one with sides 12 and 14?"
    • Agent: "The area of a rectangle with sides 12 and 14 is 168 square units."
  • Output
    • User: "What about one with sides 12 and 14?"
    • Agent: "The area of a rectangle with sides 12 and 14 is 168 square units."
Designing Agentic Systems with LangChain

Follow-up questions

{'user_input': 'What about one with sides 12 and 14?',

'agent_output': ['HumanMessage: What is the area of a rectangle with sides 
5 and 7?', 'AIMessage: The area of a rectangle with sides 5 and 7 is 35 
square units.',
'HumanMessage: What about one with sides 12 and 14?', 
'AIMessage: The area of a rectangle with sides 12 and 14 is 168 square 
units.',

'HumanMessage: What about one with sides 12 and 14?',
'AIMessage: The area of a rectangle with sides 12 and 14 is 168 square 
units.']}
Designing Agentic Systems with LangChain

Conversation history

from langchain_core.messages import 
HumanMessage, AIMessage
Designing Agentic Systems with LangChain

Conversation history

from langchain_core.messages import 
HumanMessage, AIMessage


message_history = messages["messages"]

Start of linear conversation set-up with "message history".

Designing Agentic Systems with LangChain

Conversation history

from langchain_core.messages import 
HumanMessage, AIMessage

message_history = messages["messages"]

new_query = "What about one with sides 
4 and 3?"

Conversation set-up updated with new query.

Designing Agentic Systems with LangChain

Conversation history

from langchain_core.messages import 
HumanMessage, AIMessage

message_history = messages["messages"]

new_query = "What about one with sides 
4 and 3?"


# Invoke the app with the full message history
messages = app.invoke({"messages": 
           message_history + [("human", 
           new_query)]})

Conversation set-up updated with app invocation.

Designing Agentic Systems with LangChain

Conversation history

# Extract the human and AI messages
filtered_messages = [msg for msg in 
                    messages["messages"] if 
                    isinstance(msg, 
                    (HumanMessage, 
                    AIMessage)) 
                    and msg.content.strip()]

Conversation set-up updated with message filtering.

Designing Agentic Systems with LangChain

Conversation history

# Extract the human and AI messages
filtered_messages = [msg for msg in 
                    messages["messages"] if 
                    isinstance(msg, 
                    (HumanMessage, 
                    AIMessage)) 
                    and msg.content.strip()]


# Format and print the final result
print({
    "user_input": new_query, "agent_output": 
    [f"{msg.__class__.__name__}: 
    {msg.content}" for msg in 
    filtered_messages]})

Conversation set-up updated with message printing.

Designing Agentic Systems with LangChain

Conversation history output

{'user_input': 'What about one with sides 4 and 3?',

'agent_output': ['HumanMessage: What is the area of a rectangle with sides 
14 and 4?', 'AIMessage: The area of a rectangle with sides 14 and 4 is 56 
square units.',
'HumanMessage: What about one with sides 4 and 3?', 
'AIMessage: The area of a rectangle with sides 4 and 3 is 12 square 
units.',

'HumanMessage: What about one with sides 4 and 3?',
'AIMessage: The area of a rectangle with sides 4 and 3 is 12 square 
units.']}
Designing Agentic Systems with LangChain

Let's practice!

Designing Agentic Systems with LangChain

Preparing Video For Download...