Skip to main content

Browser Use Agent

The Browser Use Agent is a built-in AI agent that can interact with web pages in your browser tabs. It reads page content, clicks elements, fills in forms, and extracts structured information.

Turn it on or off​

The browser agent is on by default. To turn it on or off, use either:

  • the "Enable browser agent" switch in the chat's Tools menu;
  • the "Browser automation agent" switch in Options → Tools → Built-in Tools.

Both switches show the same setting.

Capabilities​

The agent can perform the following actions on a tab:

  • Extract page content — read the full text content of a web page
  • Extract DOM elements — identify interactive elements (links, buttons, inputs, etc.)
  • Click elements — click buttons, links, and other interactive elements
  • Input text — fill in form fields and search boxes
  • Submit forms — simulate pressing Enter to submit forms

Tabs​

While the browser agent is on, BrowseWiz can also open new tabs in the BrowseWiz Tabs group, load pages in them, list them, and find your active tab. Each agent task works in one tab.

Tab access​

The first time BrowseWiz needs to work with your tabs, the chat shows "BrowseWiz needs permission to manage your browser tabs to do that." with a "Grant tab access" link. Click it and allow the permission in Chrome's prompt. You only need to do this once.

The agent acts freely on tabs in the BrowseWiz Tabs group. Before it acts on any other tab, the chat shows "BrowseWiz agent wants to act on domain to work on your request." with two buttons:

  • Allow access adds the tab to the BrowseWiz Tabs group. The agent keeps access to the tab until you remove it from that group.
  • Deny refuses access. The agent does not ask again for that tab until you start a new chat.

This question is asked even when approvals are bypassed. Scheduled task runs never act on tabs outside the BrowseWiz Tabs group.

Approvals​

Before the agent takes a batch of actions on a page, such as reading it, clicking, typing or pressing Enter, the chat shows a card starting with "Browser agent wants to …" that lists the actions, with two buttons:

  • Approve runs this batch of actions.
  • Deny stops the agent.

The arrow next to Approve opens more choices:

  • Auto-approve all actions on domain approves this batch and every later action on that domain until you start a new chat. It is shown only when all actions in the batch are on one domain.
  • Auto-approve all approves every later action until you start a new chat. BrowseWiz first asks you to confirm ("⚠️ Bypass all approvals this chat session?").
  • Always auto-approve on domain adds the domain to your Auto-approve domains. Like the other domain choice, it is shown only when all actions in the batch are on one domain.
  • Always auto-approve all turns on "Always bypass all approvals".

The two "Always" choices change your settings and show "Setting saved" with an "Open options" link, where you can change them back.

When no approval is asked​

The agent runs its actions without showing the card when:

  • the action is part of a scheduled task run;
  • you approved a one-off plan with "Approve & run", until you send your next message or click Stop;
  • "Always bypass all approvals" is on;
  • every page the batch acts on is on a domain in your Auto-approve domains;
  • you chose "Auto-approve all actions on domain" or "Auto-approve all" earlier in this chat.

The safety check can still ask you, or block an action, in all of these cases.

Settings​

Open Options → Tools → Built-in Tools and click "Configure" next to "Browser automation agent". The "Configure Browser Automation" dialog has:

  • Always bypass all approvals: skip all confirmation dialogs for browser actions. It is off by default. Turning it on asks you to confirm with "Enable bypass".
  • Auto-approve domains: actions on these domains are approved without asking. By default the list holds google.com and *.wikipedia.org. An entry like *.example.com matches example.com and all its subdomains. A leading www. is ignored. With an empty list, all actions need approval.

Click "Save" to keep your changes.

caution

Be careful when adding domains to the auto-approve list or turning on "Always bypass all approvals". The agent will take actions on these sites without asking for confirmation.

Safety check​

The safety check is always on and has no setting. Before every click, text input or Enter key press, BrowseWiz checks that the action fits what you asked for, or the plan you approved:

  • the action is a step toward your request;
  • it targets the element and site you meant;
  • it has no lasting effect you did not ask for, such as sending, paying, deleting or submitting something.

The check then allows the action, asks you, or blocks it.

  • Asks you. The Approve / Deny card is shown, even on an auto-approved domain, with "Always bypass all approvals" on, or during a plan you approved. The card has a "Safety check: …" line with the reason, such as "this step isn't part of what you asked for", "this may not be the element or site you meant", "this has an effect you didn't ask for", or "the safety check is unavailable". This card offers only Approve and Deny, with no auto-approve choices.
  • Blocks it. Nothing runs, the chat shows "Safety check blocked: …" with the reason, and the agent stops. A blocked action cannot be approved.

In a scheduled task run nobody can answer, so an action the check would ask about is refused. If the check cannot run, it asks you.

Example Use Cases​

  • Web research: "Search for 'best project management tools' on Google and return the top 5 results"
  • Form filling: "Fill in the contact form on this page with the following details…"
  • Data extraction: "List all article titles on this Wikipedia page"
  • Workflow automation: "Search for flights from NYC to London on this travel site"

Tips​

  • Keep tasks focused and specific — describe exactly what data you need or what action should be taken
  • The agent works best with one clear goal per task (e.g., one search query, one form to fill, one page to read)
  • Provide relevant context in your message, since the agent does not have access to previous results