AI code reviews in IDE
Get instant feedback on your code changes directly within VS Code or JetBrains IDEs.
Unlock the power of AI-driven code reviews in VS Code and all JetBrains IDEs (including IntelliJ IDEA, PyCharm, WebStorm, and more) with Bito's AI Code Review Agent. This tool provides real-time, human-like feedback on your code changes, catching common issues before you submit a pull request.
You can start using the Agent immediately—no setup is required!
The AI Code Review Agent helps you improve your code as you develop, so you don't have to wait for days to get feedback. This accelerates development cycles, boosts team productivity, and ensures higher code quality.
The AI Code Review Agent in your IDE includes a default instance called @codereview
, which Bito has preconfigured for you to start using immediately. By default, this instance doesn’t include Git information and that’s why it has basic code understanding.
To unlock more comprehensive AI code reviews with deeper insights into your entire codebase, we recommend configuring the Agent with your Git information. This setup allows Bito to analyze your code changes more effectively by incorporating relevant code blocks from your entire remote repository as context.
You need a Bito 10X Developer paid plan to get started. For more information about costs, please visit our Pricing Page.
Prerequisites
A workspace subscribed to the Bito 10X Developer Plan. Read documentation on how to upgrade.
The root of your project must be a Git repository and opened in the supported IDE.
Configure the default
@codereview
Agent instance by providing Git information, including the Git provider name (e.g., GitHub, GitLab, Bitbucket) and a Git access token.
How to configure the default @codereview
Agent instance?
@codereview
Agent instance?In the Bito panel, click the menu icon (three horizontal lines) in the top-right corner.
Click on Configure my agents. If a popup appears asking if you want to open this link, simply click Open.
You will be redirected to the Bito Cloud website. Select your desired Git provider from the dropdown menu, then enter the Git Access Token. For detailed instructions on configuring the Agent instance for different Git providers, such as GitHub, GitLab, and Bitbucket, refer to this documentation.
Below is a video guide on how to set up a Git Access Token. You will need to enter this token in Bito Cloud when configuring the Agent instance.
How to use the Agent in IDE?
Open the Bito IDE extension.
Login to your workspace subscribed to the Bito 10X Developer Plan.
Type
@
in the chat box to open a menu listing all supported actions.To start a new code review session, choose
@codereview
under MY AGENTS. This default Agent instance is available in all 10X Developer Plan workspaces and requires no setup on your part.
If you have other AI Code Review Agent instances set up in Bito Cloud, you can select them using @<instance_name>
. These Agent instances will be displayed under the TEAM AGENTS section in the popup menu.
Execute agent actions with
/
slash commands. The following commands are supported in the Bito chat box:/review
: Performs a code review. See below for details./help
: Displays help content for the agent.
Note: If the default @codereview
Agent instance is not configured with Git information, you will receive a warning popup prompting you to either configure the instance or proceed without configuration. If you choose to continue without configuring the default Agent, it will only provide basic code review capabilities.
For the best quality code reviews, we recommend configuring the Agent with your Git information.
Performing a code review
To initiate a code review, type /review
in the Bito chat box and submit. The AI Code Review Agent will analyze the code changes in your current workspace and provide feedback, including a list of issues and suggested fixes. By default, this command reviews the differences between your local changes that are currently in your working directory but not yet staged for commit, and the remote branch.
While the code review runs in the background, feel free to continue working on other tasks. Bito will notify you directly within your IDE once the review is complete. This process generally takes only a few minutes.
/review command parameters
You can modify the /review
command with additional parameters:
/review #localchanges
: Reviews the differences between your local changes that are currently in your working directory but not yet staged for commit, and the remote branch. This is the default behavior of the/review
command if no additional parameter is specified./review #stagedchanges
: Reviews the differences between your staged changes (ready for commit) and the remote branch.
Currently, the AI Code Review Agent operates in a non-interactive, command-driven mode. In the future, we plan to introduce interactive/chat-based agents for a more dynamic and engaging code review experience.
Reviewing the feedback
Once the AI code review is complete, you’ll receive a notification in the IDE. You can view the feedback in the Bito Panel, which includes a list of issues and their fixes. Each item will contain the following details:
File name: The file containing the issue.
Importance indicator: Indicates the importance of the issue.
Line numbers: Lines that contain the issue.
Issue description: Description of the identified issue.
Code suggestion: The AI-provided fix for the issue. Collapsed by default, click "Show" to view.
You can click on any specific identified issue to open its suggested code in the diff view.
Code review session history
To check the past code reviews, click the Code Review button in the IDE's status bar near the bottom of the screen. Alternatively, you can click the Session history icon located at the top-right corner of the Bito Panel. This opens the Session history tab within the Bito Panel, which contains a list of all session types (such as chat, code review, etc.). Here, you'll find a list of previous code reviews with their status.
The available code review statuses are:
In Progress: Indicates that the review is currently running.
Complete: Indicates that the review has finished successfully, with feedback already posted in the IDE.
Error: Indicates that the review encountered an issue and couldn't be completed.
To view the code review and AI suggestions, you can open any listed code review by clicking on it from the Session history tab.
Last updated