{ "description": "Use browser actions such as click, fill, keyboard input, and scroll to complete web tasks. BrowserGym returns task goals and page observations for browser-agent training and evaluation.", "tags": ["openenv", "rl-environment", "browsergym", "web-automation", "browser-actions"], "representative_queries": [ "find a browser environment for web navigation tasks", "train a web agent on form entry and button interaction", "find an OpenEnv adapter for MiniWoB browser tasks" ] }