Why agents add autonomy to LLM apps in LangChain - Test Your Understanding

Practice - 5 Tasks

1fill in blank

easy

Complete the code to create an agent that uses a language model and tools.

LangChain

from langchain.agents import initialize_agent
from langchain.llms import OpenAI
from langchain.tools import Tool

llm = OpenAI(temperature=0)
tools = [Tool(name="Search", func=search_function)]
agent = initialize_agent(tools, llm, agent_type=[1])

Drag options to blanks, or click blank then click option'

A"zero-shot-react-description"

B"simple-agent"

C"chat-agent"

D"tool-agent"

Attempts:

3 left

💡 Hint

Common Mistakes

Using an invalid or unsupported agent_type string.

Confusing agent_type with tool names.

The initialize_agent function requires an agent_type string. "zero-shot-react-description" is a common agent type that adds autonomy by letting the agent decide which tools to use.

2fill in blank

medium

Complete the code to add a tool that the agent can use to fetch current weather.

LangChain

def get_weather(location: str) -> str:
    # Imagine this calls a weather API
    return f"Weather in {location} is sunny"

weather_tool = Tool(name="Weather", func=[1], description="Get current weather")

Drag options to blanks, or click blank then click option'

Aweather_api_call

Bfetch_weather

Cget_weather

Dget_temperature

Attempts:

3 left

💡 Hint

Common Mistakes

Using a function name that is not defined.

Confusing the tool's name with the function name.

The tool's function must be the actual function defined to get weather, which is get_weather.

3fill in blank

hard

Fix the error in the agent call to run a query with tool usage.

LangChain

response = agent.run(input=[1])

Drag options to blanks, or click blank then click option'

Aquery about weather

B"query about weather"

Cinput_text

Dagent_input

Attempts:

3 left

💡 Hint

Common Mistakes

Passing a bare word without quotes causing a NameError.

Confusing variable names with string literals.

The input to agent.run must be a string, so it needs quotes around the query text.

4fill in blank

hard

Fill both blanks to create a tool that calls a calculator function and add it to the tools list.

LangChain

def calculator(query: str) -> str:
    # Simple calculator logic
    return str(eval(query))

calc_tool = Tool(name=[1], func=[2], description="Perform calculations")
tools = [calc_tool]

Drag options to blanks, or click blank then click option'

A"Calculator"

Bcalculator

C"Calc"

Dcalc_function

Attempts:

3 left

💡 Hint

Common Mistakes

Using a function name as a string.

Using a name without quotes.

The tool name should be a string like "Calculator" and the function should be the actual calculator function.

5fill in blank

hard

Fill all three blanks to create an agent that uses both weather and calculator tools with zero-shot reasoning.

LangChain

weather_tool = Tool(name=[1], func=get_weather, description="Get current weather")
calc_tool = Tool(name=[2], func=calculator, description="Perform calculations")
tools = [weather_tool, calc_tool]
agent = initialize_agent(tools, llm, agent_type=[3])

Drag options to blanks, or click blank then click option'

A"Weather"

B"Calculator"

C"zero-shot-react-description"

D"simple-agent"

Attempts:

3 left

💡 Hint

Common Mistakes

Mixing up tool names or agent types.

Using unsupported agent types.

The tools must be named "Weather" and "Calculator" matching their functions, and the agent type for autonomy is "zero-shot-react-description".