$ loading_
Automate multi-step tasks in a real Chrome browser with live view.
This MCP tool comes from an official registry and has been updated within the last year, which lowers overall concern. However, it is closed-source, connects to a remote service, and has browser automation/execution capabilities, so the overall posture is best rated as caution rather than high risk.
The materials explicitly state that no keys or environment variables are required, and there is no indication that API keys, OAuth tokens, or local sensitive credentials must be provided. Based on the available facts, the direct credential exposure surface appears limited.
The tool is declared to connect to the remote endpoint api.webrun.ai. Its described features—running tasks in a real browser, persistent environments, and live view—typically imply that task content, page data, or session-related information may be transmitted through that service. The endpoint is identified and relevant to the stated function, but there is still a data egress surface.
The system flags explicitly indicate executes-code, and the description says it runs multi-step tasks in a real Chrome browser, which implies the ability to launch a local browser process and perform automation actions. This is a normal capability for this tool category, and the current materials do not show abnormal system permissions beyond the stated function.
Given the browser automation, persistent environments, and human takeover features, the tool likely has access to browser-session data such as page content, cookies/login state, or user data handled during automation. The materials do not explicitly claim broad local file read/write permissions, so caution is more appropriate than high risk.
Positive factors include distribution through an official registry and updates within the last year. However, there is no open-source repository, no declared license, the README is absent, and community adoption is very low, which limits auditability. Overall, the source is not obviously deceptive or suspicious, but supply-chain transparency is limited and warrants caution.
Copy the install command and let the AI configure it · recommended for beginners
Please install the "WebRun Browser Automation" MCP server from askskill: Run: claude mcp add --transport http 'ai-webrun-webrun' 'https://api.webrun.ai/mcp'
Developers or product managers can use it to execute multi-step actions inside Chrome when tasks depend on real web interactions. It fits cases that require a full browser environment rather than simple HTTP requests.
When a web task spans multiple steps and needs context preserved, the persistent environment helps continue execution. This reduces repeated logins or repeated setup steps.
For workflows that may require human judgment or manual actions, users can automate most steps and take over when needed. It is useful for browser processes that cannot be fully unattended.
It is used to run multi-step tasks in a real Chrome browser. Known features include persistent environments, live view, and human takeover.
Yes. The original description explicitly mentions live view, which means users can monitor the browser automation in real time.
The provided material does not include installation steps or dependency details. See the source repository for exact setup requirements.
Automate web page workflows, login checks, tab handling, and recovery steps.
Automate Browserbase tasks via Rube MCP and check latest tool schemas first.
Control your real Chrome to automate web tasks, testing, and data collection.
Automate browser tasks for AI agents with parallel sessions and human handoff.
Control a real Chrome browser for authenticated automation and resilient page workflows.
Automate Browserhub browser tasks via Rube MCP using current tool schemas.