Using Bolt
Please review the content below carefully. If it does not resolve your issue, we encourage you to explore our Education & Tutorials section before reaching out to Support. If you need us, we’re here to assist you and will do our best to help!
Bolt works best in Chrome and Chromium browsers. Make sure you are developing in one of these browsers. You can read more here.
Some browser extensions, such as ad blockers, can also interfere with the function of Bolt. If you are experiencing difficulty, try disabling browser extensions to see if this resolves the issue.
Currently, Bolt only supports public GitHub repositories. Append bolt.new with the URL to the public GitHub repository as follows:
https://bolt.new/[PUBLIC GITHUB REPOSITORY URL]
Click the Download button at the top-right of the Bolt page to download a ZIP file of your project code.
Click the Open in StackBlitz button at the top-right of the page to continue developing your project in the StackBlitz IDE.
Click the Deploy button at the top-right of the page. This will prompt Bolt to build and deploy the application to Netlify. Bolt provides a Netlify link to your project, along with a link to claim the project as your own.
There are two ways to open projects in Bolt:
- Use the “Open in Bolt” button located in the upper-left corner of any StackBlitz project.
- Manually enter the following URL pattern in your browser:
https://bolt.new/~/STACK_BLITZ_PROJECT_SLUG_HERE
Example:
- StackBlitz Project: https://stackblitz.com/edit/sb1-9vp4ma
- Bolt Starter: https://bolt.new/~/sb1-9vp4ma
Right-click the file(s) of interest in the Bolt code editor and select Target file to have Bolt only make changes to those file(s).
Right-click the file(s) of interest in the Bolt code editor and select Lock file to prevent Bolt from making changes to those file(s).
Rollbacks: Open your Bolt chat, scroll up to the message you want to restore your project to, and click “Rollback.”
Backups: Open your Bolt chat, click Settings > Backups, and select the timestamp you want to restore. This will create a new chat with the backup of your project without erasing the current version (a “fork”).
A myth that many people have come to believe about LLMs is that it will remember everything that you have told it during a conversation, or within one chat. The amount that LLMs process tokens is called the context window.
context window: amount of text that a large language model (LLM) can process at once
Like the image above shows, the context window is not all encompassing of all the conversation you had with Bolt, but a certain window that Bolt uses to provide a response. We have already done work to expand the context window, which has been a great improvement in performance. However, we don’t want to expand the window too large, so that tokens are not consumed too quickly.
💡 A helpful tip while using Bolt is to continually remind it of important details, even if you think they’re within the window. Be specific with your prompts!
When utilizing LLMs, it will greatly improve performance to break big changes into smaller pieces!
Try your best to focus on one area or feature at a time, instead of trying to do too many things at once. A good process to follow is:
- Make a change
- Check if the change works
- Move to the next feature
In each step, make sure that you are providing a very specific request with required context to increase chances of success.
💡 Think of AI like a helpful but forgetful assistant. Try to:
- Show Bolt what you want
- Not assume it remembers previous conversations (context window!)
- Be clear about which part you’re working on
If you find yourself stuck, ask yourself these questions:
- Have I explained which part I’m changing to Bolt?
- Am I assuming that Bolt remembers something?
- Have I described the end result I want?
If you’re still experiencing issues, reach out to us at [email protected].