​
Login

Once you have installed the web extension, click its icon in your browser’s toolbar to open the popup and log in. Make sure to use the same account associated with your subscription. If you don’t have a subscription yet, you can get one here.

​
Launch and Set Up the VSCode Extension

If the VSCode extension is not running or hasn’t been set up, you’ll see the No active connection... message in the popup. Ensure VSCode is running and the extension is configured before proceeding.

​
Enable BringYourAI on a Website

Visit your desired AI chat website and open the extension popup again.
To use BringYourAI on a new website, you first need to grant permission for the extension to access that site. Click Click here in the popup’s center and approve the requested permissions in your browser.
To turn off BringYourAI for the website you’re currently visiting, right-click the BringYourAI icon in your browser’s toolbar. Then, click the Enable BringYourAI on this domain option in the menu. This will remove the checkmark ✓ and disable the extension only on this specific site.
Once permissions are granted, the popup should display the BringYourAI is ready to use... message, indicating a successful setup for that site.
Now let’s explore the UI elements.

​
Logout

Clicking your name at the top of the popup opens a dialog with a logout button.

​
Custom Trigger Char

Click the trigger char displayed in the Type [...] to begin. message within the popup. This opens a dialog allowing you to customize the trigger char specifically for the current website. The trigger char is what you type in an input field on the AI chat website to activate the BringYourAI popover. You can either use the default global trigger char or set a site-specific one. This is helpful on websites where the default char might conflict with existing site functionality (e.g., ’@’ on ChatGPT often triggers GPT mentions, so ’#’ is the default there).

​
Global Settings

Clicking global settings at the bottom of the popup opens the global settings dialog.
Here, you can change the global trigger char used by default on all websites unless a site-specific trigger is set.

​
Auto Context

Enable ‘Auto Context’ for Rules for AI to automatically include rules defined in the VSCode extension in the first message of a new chat conversation. These rules will not be added to subsequent messages within the same conversation, ensuring they are applied once per chat session.

​
The Popover

Let’s examine the popover that appears when you type the trigger char in an input field.
Each option adds specific context to the input field:
NameDescription
Focused FileAdds the entire content of the file currently active (focused) in VSCode.
Selected Text / CursorAdds the text currently selected in VSCode. If no text is selected, it adds the code immediately surrounding your cursor, including a clear indicator for your exact cursor position. A unique ID number distinguishes multiple selections/cursors.
Visible EditorsAdds the content of all files currently visible in VSCode editors (e.g., files in split view). Does not include files open in tabs but not currently visible.
All Open TabsAdds the content of all files currently open in VSCode tabs, including those not actively visible.
ProblemsAdds all problems detected by VSCode (from the Problems Panel), including surrounding code context for each issue.
User Rules for AIAdds your custom Rules for AI defined at the User (global) level in VSCode settings.
Workspace Rules for AIAdds your custom Rules for AI defined at the Workspace level in VSCode settings.
Workspace Folder Rules for AIAdds your custom Rules for AI defined at the Folder level in VSCode settings.
File TreeAdds a textual representation showing the structure of non-ignored files and folders within the relevant workspace folder.
Full CodebaseAdds the content of all non-ignored files within the relevant workspace folder.
Browser TabsAdds the main text content of the selected browser tab, or you can click the small camera icon to attach a screenshot of that tab.
Search GitHub RepositoriesSearch for and attach entire GitHub repositories, including private ones when you’re signed in to the GitHub account that has access to them in that browser. You can also select any branch you need.
Context7 Library DocsSearch for and attach Context7 library docs. More info here!
You can search for specific files, folders, or the options above by typing your query after the trigger character.
Use the Up/Down arrow keys to navigate the options and press Enter to select one. You can also click an option with your mouse.
When you add a file to the chat and then edit it in VSCode, BringYourAI will automatically update the file content in your chat message to reflect the latest version.

​
Tokens

Most options display an estimated token count. This number represents the approximate number of tokens the selected context will add to your message. AI language models have maximum token limits per message (context window size); content exceeding this limit will be ignored by the AI. Token limits vary by model (check the specific model’s documentation). Monitoring these counts helps you stay within the model’s context window.
Token counts for individual files/folders might not appear immediately in the popover for very large workspaces, as the initial calculation can be time-consuming. However, the token count will be displayed in the context chip once an item is added to the input field.

​
Context Chips

Context added via BringYourAI appears as chips above the input field.
  • Click the x on a chip to remove that specific context from your message.
  • Clicking the chip itself allows you to preview its content.
  • The total token count for all added context chips is displayed above the chips.
  • Hovering over the chips reveals a Hide toggle, which lets you hide them in case they obstruct anything behind.
Context added via BringYourAI also gets shrunk down into context chips within your chat history messages to avoid clutter.

​
Multiple VSCode Instances or Workspace Folders

BringYourAI seamlessly handles multiple VSCode windows and multiple workspace folders within a window. No extra configuration is needed. The popover UI will group context options by VSCode instance and/or workspace folder, and search spans across all connected instances and folders.

​
Context7 Library Docs

Context7 is a large collection of LLM-optimized library documentation that also allows you to semantically search for specific topics, so you don’t have to include everything or worry about token limits. These extra settings are accessible via the small button on the right of each Context7 search result. To go even faster, you can append settings directly to your search query using colons, while having your desired docs option focused and pressing Enter. For example: @sveltekit:routing:20000 In this format, the string after the first colon is the topic, and the number after the second colon is the token limit. By default, the token limit is 10,000. To quickly attach the full documentation, just add :full to your query, like this: @sveltekit:full