The Browser Use tool gives your Agent the ability to automate complex web browsing tasks—navigating sites, clicking elements, filling forms, and extracting data—all within a live browser session you can monitor and control. Unlike standard Web Search, which retrieves information from publicly accessible pages, Browser Use is designed for interactive web tasks that require navigating multi-step interfaces, dynamic search tools, and sites that don't offer APIs, such as government databases and vendor portals.
When your Agent uses Browser Use, a side-by-side browser preview opens alongside your Chat so you can watch the Agent work in real time, review screenshots of each action, and take control at any moment to guide the process.
Browser Use works within Agents and can also be called by Workflows for more complex multi-step automations.
Enabling Browser Use
Navigate to the Agents page: Click on "Agents" in the left sidebar to access your list of Agents.
Select the Agent: Choose the Agent you want to add Browser Use to by clicking on its card or opening the ellipses (more options) menu and selecting "Edit."
Access the Tools section: In the Agent's settings, navigate to the Tools section of the "Setup" tab to configure and manage tools available to your Agent.
Add Browser Use: Click "+ Add" and select "Browser Use" to enable this tool for your Agent.
Publish your Agent: Click "Publish" in the top right corner to save and activate your changes. Browser Use will not be available in Chat until the Agent is published.
For more information on editing Agents and configuring tools, refer to the Editing an Agent article.
Using Browser Use in Chat
Start a Chat: Open a new Chat and select the Agent with Browser Use enabled from the Agent dropdown.
Send your request: Type a message that requires interactive web browsing. For example: "Search SAM.gov for active contract opportunities related to cybersecurity and summarize the top results."
Watch the Agent work: The Agent opens a live browser session and begins navigating. As it works, you'll see real-time status updates for each action in the Chat, such as "Navigate to SAM.gov" or "Clicking Search menu".
To open the live side-by-side browser view, click the status dropdown to expand the list of steps, then click the latest step. This opens the browser preview panel on the right side of your Chat, where you can watch the Agent browse in real time. During the live view, you can "Click to take control" to manually guide the browser session.
Review browser steps: Once the Agent finishes browsing, click the "Finished analyzing" dropdown in the Chat to see the full sequence of actions taken. Each step is listed with a short description of what the Agent did.
Browse screenshots: Click on any individual step in the action list to open the side-by-side browser preview panel. Use the "Previous" and "Next" buttons at the bottom to navigate through screenshots of each action. The step counter (e.g., "5 / 14") shows your position in the sequence.
Expand the preview: Click the expand icon at the top right of the browser preview panel to view the screenshot in full screen. Click it again to return to the side-by-side view.
Close the browser panel: Click the "X" at the top right of the browser preview panel to close it and return to the full Chat view.
When to use Browser Use
Browser Use is ideal for interactive web tasks that go beyond what standard web scraping or the Web Search tool can handle:
Government databases and portals: Search SAM.gov, EDGAR, court records, or other public-sector databases that require navigating multi-step search interfaces and filters.
Vendor and compliance portals: Access hard-to-reach data from vendor portals, certification databases, and regulatory sites that don't have APIs.
Dynamic search interfaces: Interact with sites that rely heavily on JavaScript, dropdowns, pagination, or form submissions to display results.
Form submissions and site audits: Automate repetitive tasks like filling out forms, submitting requests, or conducting visual reviews across multiple pages.
Competitive intelligence: Navigate competitor sites, pricing pages, and review platforms to gather structured data.
If your task only requires reading the content of a publicly accessible web page, the Web Search tool is a faster and more efficient option. Browser Use is best suited for tasks that require clicking, navigating, and interacting with a website.