| title | sidebar_label | description |
|---|---|---|
FAQs |
FAQs |
Frequently asked questions about CodeRabbit. |
Early adoption results for CodeRabbit have been highly promising, demonstrating significant accuracy in code reviews. However, it's crucial to understand that AI is an evolving field, and absolute 100% accuracy can't be guaranteed. Our technology is continuously improving, aiming for the highest possible accuracy in reviews.
CodeRabbit is designed to work with all programming languages. However, the proficiency of our AI models might vary between languages based on their popularity and the public availability of training data.
We do not store your code. The code collected at the time of the review is disposed off as soon as the review is complete. During the review, there is complete data isolation of the code being reviewed, and no one from CodeRabbit or outside has access to the code.
We train our system only using publicly available datasets. CodeRabbit never uses the proprietary code from private repositories for training purposes.
Yes, you can switch between different organizations on CodeRabbit. To do so, click on the organization name in the top-right corner of the CodeRabbit UI.
Subscription seats are tied to the specific GitHub/GitLab organization under which they are purchased and cannot be used under another organization.
CodeRabbit would by default only review new PRs or existing PRs which have a new
commit after the app is installed. You can, however, use the
@coderabbitai review command on the PR to trigger a review of existing PRs.
You need to be a GitHub/GitLab admin to add the repositories.
OpenAI cost is included as part of the subscription. You don't need to have your own OpenAI key.
CodeRabbit currently utilizes OpenAI's gpt-4-turbo and gpt-3.5-turbo. We're
researching and testing upcoming LLMs to ensure we're offering the most precise
reviews possible.
Reviews are customizable. For information, check out our Customization guide.
CodeRabbit supports most widely used languages. You can configure this in the repository settings.
CodeRabbit requests minimal access to perform code reviews and post comments on Pull Requests. On GitHub, it requires read access to metadata, code, discussions, issues, and read/write access to the pull requests. On GitLab, it requires read access to the repository, and the CodeRabbit Bot user requires a developer role.
To manage users, log in to CodeRabbit and navigate to subscriptions. You can add or remove users as needed.
To interact with CodeRabbit's bot, reply to the CodeRabbit Comment. If there are team members collaborating on pull requests, the bot stays silent by default but can be engaged by tagging @coderabbitai. This feature allows you to provide context, generate test cases, or ask for specific code suggestions, all within the context of your code lines or entire files.
Absolutely! Whether you're an individual developer or part of an organization, CodeRabbit can be your coding assistant, providing you with invaluable suggestions.
CodeRabbit reviews pull requests on the default branch by default. You can change this in the repository settings.
There are hourly rate limits for each developer per repository:
- Number of files reviewed per hour: 200
- Number of reviews : 3 back to back reviews followed by 4 reviews/hour
- Number of conversations: 25 back to back messages followed by 50 messages/hr
In-trial and open-source plans have lower rate limits than the paid plan. In all cases, we re-allow further reviews/conversations after a brief timeout.
If there are any access restrictions by domain, then you will need to add
coderabbit.ai as an allowed domain.
:::tip
If you have further questions or need additional information, please check out our Support page for more details.
:::