Skip to content

Setup

Capybara Building

As with any project, the first step is setting up your environment. Let's get started by ensuring you have a project structure ready to work with.

Important

Since this is an intermediate-level course, please ensure you've gone through the Griptape Setup - Visual Studio Code course to set up your environment. We will be starting from the code at that point.

Create a Project

Following the instructions in Griptape Setup - Visual Studio Code please:

  1. Create your project folder. Example: griptape-structure-tool
  2. Set up your virtual environment
  3. Ensure you pip install griptape python-dotenv
  4. Create a .env file with your OPENAI_API_KEY
  5. Create your app.py file with the following code:
app.py
1
2
3
from dotenv import load_dotenv

load_dotenv() # Load your environment

Create a Chatbot

In a typical setup for a TradeSchool course we’d leave the setup at this step - however, in this case we’ll take it a step further and create a Chatbot. To simplify this, we're going to use code from another course and just copy that over.

The Image Query SEO Bonanza course has a great agent we can utilize. Simply copy the code from the Chatbot section in the Code Review area and save that as your app.py.

app.py
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
from dotenv import load_dotenv

# Griptape Items
from griptape.structures import Agent
from griptape.utils import Chat

from rich import print as print  # Modifies print to use the Rich library

load_dotenv()  # Load your environment

# Create the Agent
agent = Agent(stream=True)


# Modify the Agent's response to have some color.
def formatted_response(response: str) -> None:
    print(f"[dark_cyan]{response}", end="", flush=True)


# Begin Chatting
Chat(
    agent,
    intro_text="\nWelcome to Griptape Chat!\n",
    prompt_prefix="\nYou: ",
    processing_text="\nThinking...",
    response_prefix="\nAgent: ",
    handle_output=formatted_response,  # Uses the formatted_response function
).start()

Next Steps

And there you go, your environment is all set up and you've got a chatbot all ready. In the next section, learn more about how StructureRunDrivers work.