This walkthrough will take you through your first interactions with Exponent, showing our core capabilities and how you can use them. It assumes you’ve already completed the Quickstart guide to set up Exponent.

​
Starting Exponent

If you haven’t already, start Exponent by navigating to your project directory: 
cd my-project
Then run: 
exponent run
This opens a new chat session in your browser. You’ll know Exponent is ready when you see the “Connected” status in the center of the screen, indicating that Exponent has established a connection to your local environment. Exponent web interface showing the connected status

​
Running Your First Command

Let’s start with something simple to confirm that Exponent can interact with your system. Ask Exponent to list the files in your current directory: 
List the files in this directory
Exponent will analyze your request and propose running the appropriate command (such as ls -la on Unix systems or dir on Windows). When Exponent suggests a command, you’ll see something like this: Exponent proposing a command to list files with a confirmation button Use the Ctrl/Cmd-Y shortcut shown or click “Confirm” to execute the command. Exponent will run the command locally and display the results directly in the chat.
This ability to execute commands on your behalf is one of Exponent’s key features. You no longer need to switch contexts between a chat interface and your terminal, and Exponent is able to automatically view the results of the command and take action based on them.

​
Understanding Your Codebase

Now let’s see how Exponent can string together multiple commands. Try asking: 
What does this project do? Give me a brief overview.
Exponent will search through your project files, read their contents, and provide an analysis of what your project does. It might look through README files, documentation, source code, and other files to gather this information. At each step, Exponent will show you exactly what it’s doing and why. At this stage, you are in full control of approving or requesting modifications to each step. This simple example shows Exponent’s ability to:
  1. Find relevant files in your codebase
  2. Read and understand their contents
  3. Synthesize information from multiple sources

​
Modifying Files

Let’s see how Exponent can help you create and edit files. Try asking: 
Now create a file called poem.md with a brief poem about this project
Exponent will draft and propose creating a new file.
Exponent always shows you exactly what changes it’s going to make before it makes them, so you can review file modifications before approving them.
Exponent writing a file to the filesystem When you confirm, it will write the file directly to your filesystem; you can run cat poem.md on Unix systems (or type poem.md on Windows) from your terminal, and you’ll see the file contents.

​
Keep Going

You’ve now seen the core capability of Exponent: combining command execution and file operations into a powerful agentic programming assistant. The best way to get comfortable with Exponent is to continue experimenting. Try asking Exponent to help with tasks you’re currently working on, tasks you’ve used other tools for, or just do something silly that might make you smile 😃. Improving your skill at using Exponent is a process of trial and error, and the more you use it, the more effective you’ll become. Graphic wrapping up the walkthrough

​
More Exponent

Features

Explore Exponent’s full capabilities

Exponent Shell

Use Exponent directly in your terminal

Exponent CI

Integrate Exponent into your CI/CD pipeline

​
Get in Touch

Join the Community

Connect with other Exponent users and the team

Get Support

Reach out if you have any questions or feedback