Source: github.com/browser-use/browser-use/releases/tag/1.0.7
The Python package browser-use released version 1.0.7, which changes how long-running MCP calls to browser_exec behave. Previously, MCP calls could time out at 60 seconds. Now a call still running after 50 seconds replies with the output produced so far, while the code keeps running in the cell.
Calling browser_exec again with empty code returns the rest of the output. Calls that finish within 50 seconds reply as they did before. The release also updates the README, with a Browser Use for enterprises section and a restored Cloud API key link in the Python quickstart.
Skills, the README, the CLI and MCP now ask coding agents for integration feedback.
Why it matters if you build agents
If you run agents against browser-use's MCP server, a long browser_exec call no longer has to finish inside a client's 60-second window to return something. Agents should call browser_exec again with empty code to collect the remaining output, since the cell continues running after the partial reply.
Get stories like this every weekday
One short email with the day's agent tooling news.