$ loading_
Lets LLM agents control a real Chrome browser with vision.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "Hyperion V2" yet — see the docs or source repo.
Please open the target website in Chrome, browse step by step based on the page content, extract the key fields I care about, and summarize the result in bullet points.
Returns key page information and a structured summary of the browsing process.
Please complete a set of web actions in a real Chrome browser: open the page, click the relevant buttons, fill out the form, and tell me the result of each step.
Provides the result of each step and the final page state.
Please use the live browser view to judge the current page state, continue with the next actions, and explain any issues when they occur.
Continues actions based on visual input and explains any abnormality or failure.
Researchers or content workers can let an LLM browse real websites in Chrome, track page changes, and organize findings. It fits online research tasks that require visual inspection and judgment.
When a task spans multiple pages with clicks, form filling, and navigation, this tool lets an agent operate step by step in a real browser. It suits forms, admin pages, and multi-step web flows.
When an agent needs to decide the next action from the current browser view, it can use real-time vision streaming. This is useful for interactive tasks sensitive to page state.
It is a universal MCP server that lets any LLM agent operate a real Chrome browser. The description also mentions 5 perception engines, a heartbeat mechanism, and real-time vision streaming.
It is designed for LLM agents and emphasizes real Chrome, multiple perception engines, a resilient heartbeat, and live vision streaming. That makes it better suited for browser tasks that need visual perception and continuous control.
It is known to work with MCP-capable LLM agents such as Claude, Cursor, and Cline. For exact installation and connection details, see the source repository.
Automate browser navigation, interaction, and web data extraction for online tasks.
Control Chrome with Playwright for end-to-end browser automation and iframe diagnostics.
Control Chrome via MCP for browsing, form filling, screenshots, and logs.
Let AI browse via your real Chrome for extraction and multi-step workflows.
Control a real Firefox browser for navigation, interaction, and page inspection.
Control a real Chrome browser for browsing, interaction, and page reading.