Browser Local FAQs

Question 1

How is Browser Local different from headless browser automation?

Accepted Answer

Unlike headless tools, Browser Local is designed for AI agents to interact with your *actual, visible desktop browser*. This enables tasks benefiting from a real user profile or direct visual feedback, offering a more integrated experience for AI-driven web tasks.

Question 2

What kind of tasks can AI agents perform with Browser Local?

Accepted Answer

AI agents can perform precise web interactions (navigate, click, type, scroll), extract lightweight semantic page snapshots or full text/HTML, take screenshots, and conduct comprehensive PDF text extraction from local files, URLs, or the current tab.

Question 3

What is Browser Local?

Accepted Answer

Browser Local is an MCP server that empowers AI agents like Codex and Claude Code to safely control your *local desktop* Chromium-based browsers (Brave, Chrome, Edge) for advanced web interaction and data extraction, distinct from traditional headless automation.

Question 4

Which browsers does Browser Local support?

Accepted Answer

It supports Brave, Google Chrome, Microsoft Edge, and other Chromium-based browsers running on your local desktop, allowing agents to either attach to existing sessions or launch new isolated ones.

Question 5

How does Browser Local ensure security?

Accepted Answer

Browser Local prioritizes safety by defaulting `browser_eval` (arbitrary JavaScript execution) and file uploads to disabled. It uses isolated browser profiles for new sessions and automatically cleans up tabs opened by the AI agent upon disconnection.

Browser Local

Browser Local

主要功能

使用案例

主要功能

使用案例