browser-use 0.13.13 Adds OpenAI Browser Tool Support and Cloud Fixes
The latest release of browser-use introduces an OpenAI browser-tool driver, adds default browser tools, improves integration reliability, and supports multiple browser environments.
What changed?
browser-use 0.13.13 now includes an OpenAI browser-tool driver, providing 18 default browser tools and six optional actions through the Browser Use Actor. This release also fixes two integration issues: handling temporary empty Cloud tab URLs during navigation and keeping screenshots, crops, and pointer coordinates aligned on scaled displays. Additionally, it offers support for local browsers, Browser Use Cloud, and borrowed browser sessions.
Developers working with the OpenAI Python library are advised to install the openai-tools extra with OpenAI Python 3.31 or newer. For custom HTTP clients, version 3.x of OpenAI now uses httpx2 and recommends using DefaultAsyncHttpxClient.
Why does it matter to an everyday developer?
These updates make it easier to integrate browser automation into applications that use OpenAI tools, remove bugs related to navigation and display scaling, and provide a broader set of ready-made tools for automation. The improvements help ensure that browser-based actions and data capture (such as screenshots and mouse events) are more accurate, especially in different environments or on high-DPI screens.
What can the developer do now?
Developers should upgrade to browser-use 0.13.13 to benefit from bug fixes and expanded capabilities. They can enable the OpenAI browser-tool driver by installing the openai-tools extra when using OpenAI Python 3.31+. For integrations requiring custom HTTP clients, developers should use DefaultAsyncHttpxClient with OpenAI 3.x. It is now straightforward to work with local browsers, browser cloud sessions, or external browser session borrowing, supporting a range of browser automation and tool use-cases.
