$ loading_
Automate real browser tasks tab by tab while keeping the current session.
This tool exposes Playwright browser automation as MCP capabilities, allowing local control of a real browser while preserving the user's active session; these are typical high-privilege browser automation characteristics and fit caution rather than high risk. Supply-chain posture needs extra verification because it is listed in a third-party registry, has an auditable open-source repo, but no declared license, low adoption, and unknown maintenance status.
The material states that no extra keys or environment variables are required, which reduces explicit secret-handling risk; however, it 'preserves the user's active session,' meaning it may leverage existing browser logins to access authenticated sites, creating indirect session-abuse risk if misused.
No fixed remote endpoint is declared, but the tool drives a real browser for navigation and form filling, so network access depends on the websites opened by the user/agent; page inputs, session context, or interaction data may be sent to destination sites, but the material shows no clear red flag of exfiltration to unknown backend services.
The objective checks indicate it executes code, and the description shows it uses Playwright to control real local browser tabs; this is a normal browser-automation/MCP capability, implying it can launch browser processes and perform automation locally, with no evidence in the material of system privileges beyond its stated purpose.
By operating a real browser and preserving the active session, the tool can typically access page content, form fields, and data visible under existing cookies/session state, and may interact with logged-in resources; the material does not state local file read/write behavior, and no clearly excessive data access is described, but access within the browser session is broad.
A positive factor is that there is an auditable open-source repository; however, it comes via a third-party registry, has no declared license, 0 stars, unknown maintenance, and no README content here, which reduces assessability. The current facts are not enough for a high-risk rating, but the repo contents, releases, and dependencies should be verified before installation.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "mcp-browser" yet — see the docs or source repo.
Open the target site's signup page and, using my current session, fill in the name, email, company, and notes fields step by step, then pause for my confirmation before submitting.
The AI controls the browser to complete the form and pauses before final submission.
Open the orders page from the past week in my logged-in dashboard, review each page, and compile order numbers, customer names, and statuses into a clear list.
A structured list of the extracted information for analysis or reporting.
Use the browser to visit the product settings page and follow this flow: open settings, change notification options, save, and verify the success message, while logging each step's result.
The requested workflow is completed, with a step-by-step execution log and outcome summary.
Let AI control a real browser for navigation, forms, and web tasks.
Automate browser tasks, capture console logs, and take screenshots for web workflows.
Control Chrome with Playwright for end-to-end browser automation and iframe diagnostics.
Control a browser with AI for automation, extraction, interception, and screenshots.
Automate browser navigation, interaction, and content extraction with Playwright tools.
Let AI control a browser for web tasks, extraction, and testing.