The BrowserAct MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your BrowserAct account. It provides structured and secure access to your hosted browser workflows, templates, tasks, and proxy regions, so your agent can discover workflows, start browser tasks, track progress, resume paused tasks, and cancel running tasks on your behalf.
- Workflow discovery: Have your agent list available custom workflows and review their details and required inputs before choosing one to run.
- Template search and review: Let the agent search official workflow templates and inspect their details and expected inputs before starting a task.
- Hosted browser task execution: Direct your agent to run custom workflows or official templates with the inputs needed for your browser automation job.
- Task monitoring and control: Instruct your agent to list tasks, check their status and progress, review outputs or failures, resume paused work, or cancel a running task.
- Proxy region selection: Have your agent view supported proxy regions and choose an available location for an official template run.