Browser Use
A tool-using browser agent available as a local library/CLI and a hosted cloud service. The agent decides browser actions from task instructions and a configured model.
Utility: Browser automation; kind: agent.
Stars: 119k+; Starred in September: 5k+.
Metrics updated: 31 Oct 2026. Profile reviewed: 31 Oct 2026.
Utilities
- Navigate websites and carry out reviewed multi-step browser tasks.
- Integrate browser actions into an agent application.
Who it is for
- Developers and technical users building browser-based agent workflows.
Requirements
- Follow the current Python 3.12 quickstart or supported CLI/library instructions.
- Provide a supported model and local browser, or cloud credentials for hosted execution.
Limitations
- Dynamic sites, login state, and anti-automation controls can interrupt tasks.
- Browser actions can submit forms or change account data; use narrow permissions and approvals.
Cost and privacy notes
- Local model/browser resources and hosted agent/browser usage are different cost paths; check current pricing.
- Browser profiles, cookies, screenshots, and page content may reach the selected model or cloud service.
Illustrative example (untested)
Illustrative, not tested: collect public event dates from a website without logging in or submitting forms.
License
MIT for repository code; hosted services have separate terms.