Hey, thanks for the report.
The Chrome option was intentionally removed in recent versions for security reasons. It was an internal Playwright MCP build that we’ve deprecated. So this isn’t a bug, it’s a product change.
Alternatives for your workflow:
- Browser Tab (built-in browser)
- Settings > Tools & MCP > Browser Automation > Browser Tab
- Same capabilities: navigation, clicks, console, network
- Difference: the browser runs inside Cursor, not in a separate window
- Docs: Browser | Cursor Docs
- Microsoft Playwright MCP (if you need an external Chrome)
- GitHub: GitHub - microsoft/playwright-mcp: Playwright MCP server
- Add it to your MCP servers config manually
- Full browser control, better supported
Other users confirm Playwright MCP works as a replacement: No access to browser automation
Try Browser Tab first. If a separate window is critical for your use case, Playwright MCP should solve it. Let me know if that doesn’t work for you.