Computer Use & Browser Automation

Browser automation gives your agent the ability to browse websites, interact with web applications, extract data, fill forms, and take screenshots — using direct browser actions and Computer Use (vision-based UI control) when needed.

---

What your agent can do

• Navigate to URLs and interact with pages
• Click, type, scroll, and hover via Computer Use (it reads the screen)
• Extract structured data from web pages
• Take screenshots and describe what’s on screen
• Download and process files in the sandbox
• Run terminal commands in an isolated sandbox

---

How it works

Each agent shares one sandbox session per organization (keyed by agent). The session persists across runs so files and environment state carry over:

1. When a run starts, the agent’s sandbox session starts in the background, or the one from its earlier runs is reused
2. Browser navigates to the target URL
3. A screenshot and page context are sent back to the agent
4. Agent chooses the next action (navigate, Computer Use delegation, extract data)
5. Steps repeat until the task is complete
6. The sandbox session stays available for the next run (not destroyed after each task)

Note

The cloud sandbox runs in an isolated container and cannot reach your local machine. Workspace files are restricted to /workspace. On paid plans an Internal agent can run on your own machine instead; see Run agents on your machine.

Note

On the Free plan, the agent’s computer can use the internet only over the web (HTTPS and HTTP, ports 443 and 80). Other outbound connections, such as to email servers or over SSH, are blocked. Paid plans have no such limit.

---

Enabling browser automation

1. Open your agent in the Console
2. Go to Capabilities
3. Check that Browser is on (it is by default)
4. Choose Save changes

Computer Use comes with the browser and has no switch of its own. An Internal agent always has its sandbox, so terminal and file access don’t need a switch either. Customer-facing agents don’t get a browser, sandbox or Computer Use.

---

Use cases

Web Data Extraction

Task: Go to https://news.ycombinator.com/front and extract the top 10 stories.
For each story, get:
- Title
- URL
- Number of points
- Number of comments
Return the results as a markdown table.

Form Automation

Task: Go to https://forms.example.com/onboarding and fill in the form:
- Company name: Acme Corp
- Contact: John Smith
- Email: john@acme.com
Submit the form and confirm it was successful.

UI Testing

Task: Test the login flow on https://app.example.com:
1. Navigate to the login page
2. Log in with the test credentials provided in my message
3. Verify you reach the dashboard
4. Take a screenshot of the dashboard
5. Report any errors encountered

Monitoring

Task: Take a screenshot of our status page at https://status.example.com.
If any service shows a non-green status, send an alert to #ops-alerts on Slack
with the screenshot attached.

---

Tips for reliable automation

Be explicit about UI elements

Instead of “click the login button”, say “click the button that says ‘Sign In’ near the top right of the page.”

Add error handling

“If the page takes more than 10 seconds to load, try refreshing once. If it still fails, report back with the URL and a screenshot.”

Test in the panel first

Test browser tasks in the Try it pane on the agent’s page before you put them in a Loop.

Never hardcode credentials

Pass credentials in your message or store them in integration connections — do not put secrets in agent instructions.

---

Compute usage

Each plan includes computer hours each calendar month. Sandbox and browser time within them costs no credits, and each Computer Use session costs credits. Credits cannot buy more hours. When your team has used them all, a new computer does not start until the next calendar month or until you upgrade your plan. Check computer time in Usage (under Operate in the Console) and credits in Settings → Billing & usage. See Pricing & Credits → for the numbers.