$ loading_
Let Claude control a real Chrome browser through CDP actions.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "Controlium" yet — see the docs or source repo.
Using Controlium, open the target webpage, click the login button, enter the test username and password, submit the form, and take a screenshot so I can verify the result.
The AI controls a real browser to navigate, click, type, and capture a screenshot, then returns the result.
Open this page, scroll to the main content area, read the key visible information on the page, and save a screenshot of the current view.
The AI scrolls the page and captures a screenshot, then summarizes the visible content from the browser.
After opening this webpage, run a JavaScript snippet, read the text from a specified element, and tell me the execution result.
The AI executes JavaScript in the real browser context and returns the script output or extracted page data.
Developers or product managers can have AI run login, navigation, and click flows in a real browser to quickly verify expected behavior. It is useful when visible cursor movement and real page feedback matter.
Researchers can ask the AI to open pages, scroll through content, and take screenshots to help document key information. It fits tasks that need both inspection and visual evidence.
Developers can have the AI execute JavaScript in the browser to read page elements or inspect frontend state. This is useful for lightweight debugging in a real page context.
It is a tool made of a Chrome extension and an MCP bridge that lets Claude control a real browser through CDP. It supports navigation, clicking, typing, scrolling, screenshots, and JavaScript execution.
Based on the description, it can navigate pages, click, type, scroll, take screenshots, and execute JavaScript. It also mentions a visible cursor and bringing tabs to the front.
Known prerequisites include a Chrome extension and an MCP bridge, with communication through CDP to a real browser. More specific installation and runtime requirements are not provided here; see the source repository.
Control a real Chrome browser for browsing, interaction, and page reading.
Debug web apps through Chrome DevTools using natural language commands.
Control Chrome via MCP for browsing, form filling, screenshots, and logs.
Monitor and control your computer with real-time system, file, code, and web tools.
Gives Claude Code real browser control for automation, testing, and screenshots.
Control your real Chrome to automate web tasks, testing, and data collection.