Browse websites, use saved logins, and review or take over your agent’s browser.
The agent browser lets your agent browse websites, sign in with saved logins, fill forms, and download files. It runs Chromium on the agent’s Computer and supports sites that need JavaScript or a signed-in session.You can watch the browser live from the chat, take over when the agent needs a human, and replay every browser step after the task finishes.
Open pages, click, type, scroll, switch tabs, and read what is on screen, including JavaScript-heavy pages.
Sign in without seeing passwords
Type saved logins and 2FA codes into a site. The agent never sees the password. See Browser Logins and Profiles.
Stay signed in across tasks
Sign-ins are saved to a browser profile at the end of each turn and restored on the next task.
Ask you for help
When it hits a login, 2FA code, CAPTCHA, or purchase confirmation, the agent pauses and hands the browser to you.
Show its work
Watch the browser live in the side panel, and replay a recording of every browser step afterwards.
Browse from another country
Route traffic through a residential proxy so sites see a visitor from the country you pick.
The browser can fill forms embedded from other sites, such as Persona identity checks, Stripe or Shopify card fields, and sign-in widgets. The agent can type card numbers and codes you hand over into a payment frame on the page, but it types passwords only into a frame from the password’s own site.
Use the browser for sites and tasks no connector covers. When a connector covers the app (Gmail, Salesforce, Notion, and so on), it is faster and more reliable than clicking through the site.
Your organization hasn’t given your role access to the browser yet. Reach out to your organization admin and ask them to add the Agent browser feature to your role. Until they do, agents you run work without the browser.
When the agent starts using the browser, a Browser bar appears just above the message box, with a small preview of the page, the number of browser steps, and the elapsed time. The panel does not open on its own. Click the bar to open the Browser panel on the right side of the chat, and close the panel with its X. The bar stays available while the browser is reachable, so you can reopen the panel later or bring it back in front of a tool call or file. The one exception is a handoff: when the agent asks for your help, the panel opens for you.
The panel has two modes, switched from the pill at the bottom of the panel.
Live
Replay
Live streams the agent’s browser as it works, with its real tab strip, Back, Forward, Reload, and an address bar.
While the agent is driving, the page has a blue glow.
You can click and type directly in the page. Your clicks and typing go to the real browser.
Type a URL or search terms in the address bar. Anything that is not a web address becomes a Google search.
Open a new tab with + and switch tabs from the tab strip.
If someone else in the chat is using the browser, you see a label with their name, such as Max is using the browser.
The blue glow means the agent is driving.
Replay · N steps plays back a recording of every browser step in the chat, as one timeline.
Step back and forward between browser steps, play or pause, scrub, and change the playback speed.
The tab strip and address bar replay with the clip, so you can see which page each step was on.
Screenshots the agent took are marked on the timeline.
Open the replay later from Browser replay under the agent’s message.
Only one driver at a time. While a person is controlling the browser, the agent cannot run browser steps on the same page. During a handoff, only the person the browser was handed to can control it.
Every browser step is recorded. Recordings are attached when the turn finishes, so a running or paused turn never shows a half-finished clip. A very long step is cut off at the frame limit and marked Truncated at the frame limit.Recordings from a subagent stay in that subagent’s own chat; they are not copied into the parent chat.
The agent hands the browser to you when it reaches something only you can do: a sign-in it has no login for, a 2FA or one-time code, a CAPTCHA it could not pass, an SSO button, approving on your phone, or confirming a purchase. This is a browser handoff. The agent pauses until you answer.For an API key, token, or other non-login secret, the agent can reuse an available saved secret instead of asking you to enter it again.
Use the browser in the side panel, then continue. The agent is paused until you do.
Instructions
What the agent needs, written by the agent.
Fields
Up to six inputs, such as a username, password, or one-time code. Use the options below to enter a sensitive value or select a saved secret.
Only used on <host>
Gumloop locks sensitive values, including saved secrets you select, to the site the browser was on when it asked. For example, the agent can type a password requested on app.acme.com on login.acme.com, but never anywhere else.
Save login for <site>
Saves a newly entered username and password to your Vault so the agent can sign in next time without asking. Pick Only me or My team when both are offered.
Skip
Decline. The agent continues without your help.
Open browser / Continue
Open the live panel if it is closed, then continue once you have entered values or finished in the browser.
For a sensitive field, choose New to enter a value or Existing > Select a secret to reuse a personal or team Vault secret you can access. The picker does not include saved website logins or connector credentials. Selecting a secret for the password does not offer Save login.The agent types the username, password, and 2FA code you provide, including across both pages of an email-first sign-in. Gumloop keeps the values until they are typed, for up to 30 minutes, so a later turn can still use them. When a page has a Google, Microsoft, or Okta sign-in button, the agent presses it first instead of asking you on that page.Click Continue when the fields are ready. Gumloop types sensitive values only on the site shown in the card, without adding them to the chat transcript. Password fields are masked in the live view, recordings, and screenshots.
A resolved handoff that saved the login.
Use the live panel to sign in, enter the code, or pass the check, then click Continue. The agent looks at the page and picks up from there. Use this for SSO buttons, CAPTCHAs, phone approvals, and purchase confirmations, where there is nothing to type.
When the chat runs in Slack, the handoff posts to the thread with buttons:
Handoff
Buttons
With fields
Enter the details (opens a form), I did it in the browser, I couldn’t do it
Without fields
Open in Gumloop, I did it in the browser, I couldn’t do it
Sensitive fields in the Slack form show the host they can be typed on. You can also open the browser in Gumloop, complete the step there, then hand it back from the thread.
Open the secure link in Gumball’s text. A handoff with fields opens an answer page; a handoff without fields opens the live browser. Complete the requested step there, then hand it back to the agent. The continuation comes back by text.Do not text passwords, API keys, or one-time codes. Provide sensitive values on the secure page, which shows the host they can be used on. Replying approve is not a browser handoff answer.
The card resolves to one of: Completed in the browser, Could not be completed in the browser, or Saved the login for <site> as <NAME> and sent it to the agent.
Only the person whose message the agent is working on can answer its handoff. Everyone else in the chat sees the card as waiting for that person.
Choose Auto or a country from Proxy in the Agent tab’s Browser section. Proxied traffic exits through a residential connection in that country, so sites see a visitor from there.
Option
Where traffic exits
Off
Directly from the agent’s computer, without a residential proxy.
Auto
The country associated with the person running the task’s saved timezone. If that country cannot be determined or supported, the exit is in the United States.
Country name
The country you select.
When enabled in the configuration panel, the browser defaults to Auto. Existing agents keep their saved proxy choice. Gumball uses a proxy based on your saved timezone without per-agent setup.
If a site refuses the proxy, the agent can switch that task to a direct connection and later switch back to its configured proxy. This requires a configured proxy and does not change the agent’s saved setting.
Switching the task’s connection restarts the browser. Open tabs and in-page state are lost; sign-ins and the current device choice are kept.
Choose Mobile from Default device in the Agent tab’s Browser section to start each task with a mobile browser. Desktop is the default.
Default device in the agent's Browser section.
Device
Browser behavior
Desktop
Desktop Chromium.
Mobile
A mobile viewport, touchscreen input, and an Android Chrome user agent, emulating a Pixel 7.
Mobile mode tests websites, not native mobile apps.You can also ask the agent to switch devices during a task, for example, “Check this page on mobile, then compare its desktop layout.” The switch keeps your tabs and sign-ins and applies to all tabs in that task. It does not change Default device for future tasks.
The browser remembers which websites it is signed in to with a browser profile. At the end of each turn, Gumloop saves any new sign-ins to the profile, and the next task starts already signed in, even in a brand-new chat.
After the handoff, the agent is signed in and reports the saved login. Later chats on this agent start signed in.
When a saved session has expired, the agent signs back in with a saved login and the end-of-turn save keeps the new session. For profile ownership, importing your own browser’s sign-ins, and 1Password, see Browser Logins and Profiles.
The browser has no separate charge. Browser steps aren’t billed as tool calls, and the proxy, recordings, and live view are free. A browser task is billed like any other agent chat:
Compute: 5 credits per minute the agent spends working in the browser.
Chat & Reasoning: the model reading pages and screenshots.
Time spent waiting on you during a handoff isn’t billed.
No. Login passwords, 2FA setup keys, and sensitive handoff answers are never shown to the agent and are not stored in the sandbox environment. The agent refers to them by name, and Gumloop types them in for it. 2FA codes are generated by Gumloop at the moment of typing, so the setup key never enters the sandbox. The username is not secret: the agent can read it.
Can a page trick the agent into typing my password somewhere else?
Every secure fill is checked against the login’s site before anything is typed: the scheme, port, and host of the frame the field lives in must match. A mismatch types nothing and tells the agent to ask you instead. Handoff answers are locked to the exact host they were asked on, and the card shows you that host before you type.
Can someone see the password in the live view or recording?
No. Password and code fields are masked before and after typing, including if the site has a show-password toggle, and a screenshot is skipped if a typed field cannot be masked. Values are also kept out of chat transcripts and logs.
What should admins assume about bound logins?
These protections defend against a web page steering the agent. They are not a hard boundary against an agent that sets out to extract a value from its own sandbox. Treat every login an agent can type as usable by that agent, and bind only the logins it needs. For tighter control, restrict Agent browser with Custom Roles.
Check three things: the Browser section is turned on and saved, your role allows Agent browser, and the task actually needs a browser. If a connector can do the job, the agent may use it instead. To force the browser, ask explicitly, for example “Use your browser to open …”.
I turned the browser on, but the agent still can't use it. Why?
If your organization uses Custom Roles, Agent browser is denied by default and is checked for whoever runs the agent, not whoever configured it. A teammate whose role lacks it gets an agent without the browser. See Why does it say “Turned off for your role”? for what to ask your admin and how they turn it on.
Can I take over the browser while the agent is working?
Yes. Click into the Live view and use the page. While you hold control, the agent cannot run browser steps on that page, so it is best to take over when the agent asks with a handoff.
Why did my open tabs disappear?
The browser restarts when the proxy setting changes or when it switches to a different browser profile. Tabs close; sign-ins in the profile are kept.
The agent signed in, but the next chat is signed out. Why?
The usual causes:
The chat was incognito. Incognito chats never save or restore sign-ins.
Someone else ran it. With the default User-owned profile, each person has their own sign-ins, so a teammate’s run starts from their profile, not yours.
The site signed the browser out. Sessions expire. Give the agent a login so it can sign back in on its own.
Yes. Choose Mobile under Default device in the Browser section, or ask the agent to switch during the task. It uses a mobile viewport, Android Chrome user agent, and touch emulation.
What happens when a site shows an alert, confirmation, or text prompt?
Alerts are dismissed automatically. The agent can answer confirmation dialogs and text prompts, or ask you for a browser handoff when the decision needs your input.
Can the agent get past CAPTCHAs?
If the agent cannot pass a CAPTCHA, it asks you to solve it in the live browser.
