Automate browser profiles with an API & MCP

Running accounts at scale means doing the repetitive work with code, not clicks. Alias Browser exposes two automation surfaces — a local REST API and a Model Context Protocol (MCP) server — so scripts and AI agents can create, launch, and drive isolated profiles programmatically.

How can I use an API to automatically create and manage browser profiles in an anti-detect browser?

The pattern is the same across tools: a local API manages the profiles, and a standard automation protocol drives the page. With Alias Browser:

• Manage profiles via REST — list profiles, launch a profile, launch it with an automation channel, and stop it.
• Drive the page — once a profile is running with automation enabled, connect over the standard protocol to navigate, read content, click, type, and screenshot.

Because every profile keeps its own fingerprint, cookies, and proxy, you can fan out across many identities in parallel while keeping each one isolated.

What is MCP and how does it relate to browser automation?

MCP (Model Context Protocol) is an open standard that lets AI assistants call external tools in a structured, predictable way. Instead of writing custom integration code, you point an MCP-capable agent (for example, an AI coding or research assistant) at the Alias MCP server, and it gains a set of browser tools:

• navigate — open a URL in a profile.
• get_page_content — read the rendered page text/DOM.
• evaluate_javascript — run JS in the page context.
• click_by_index / type_by_index — interact with elements.
• screenshot — capture the current view.

This lets an AI agent operate a real, isolated anti-detect profile end to end — useful for research, monitoring, QA, and agentic workflows.

Which engine supports automation?

Automation in Alias Browser runs on its Firefox-based engine over the standard WebDriver BiDi protocol, which supports the complete automation suite. The setup is simply: create or pick a Firefox-based profile, launch it with automation enabled, and point your REST client or MCP agent at it. Chromium-based profiles remain available for normal interactive browsing.

A typical workflow

1. Create the profiles you need (each gets its own fingerprint + proxy).
2. Launch a profile with the automation channel via the REST API.
3. Connect your script or MCP agent and run navigate / read / click / type.
4. Stop the profile when finished; repeat across profiles in parallel.