browser-control

Give your AI Agent the power to control Safari on macOS. No extensions, no separate browser.

Chrome browser extension for Hermes Agent — exact sessions, user-selected tabs, and local-first browser control.

Control your browser from a Home Assistant lovelace card: full screen, disable screen lock, zoom, reload page...

Take the reins of your real browser

🦞 Zero-config CLI to control a real Chromium browser from the shell — built for AI agents & scripts. Navigate, click, type, screenshot, extract page structure, Gemini vision analysis. Playwright-based. 零配置命令行浏览器控制工具，让 AI Agent 用简单命令操作浏览器。

Open-source CLI bridge for shell-capable AI agents to work in existing Chrome tabs. Local Node server, no third-party runtime dependencies, scoped snapshots, and visible action feedback.

Local MCP server and Chrome browser control runtime that lets AI agents use your existing browser session.

Distributed AI agent management platform. Run, manage, and coordinate multiple AI agents across your infrastructure from a single dashboard.

Chrome and Edge browser control for Pi, Codex and DSH. Drives your real browser profile, login state and tabs through an MV3 extension and a local loopback Bridge - native CDP, AX-first semantics, no separate browser.

MCP server that drives the Chrome/Edge tabs you already have open and signed in — local extension pairing, no remote debugging port, no separate automation browser. Tools for AX snapshots, native click/type/fill/hover/drag/scroll, batch steps, waits, navigation, dialogs, upload, screenshots, JS evaluate and console/network logs.

Low latency web streaming and motion-triggered surveillance / security / wildlife camera control application for Raspberry Pi that supports some of the basic features of RPI_Cam_Web_Interface.

Live semantic browser Truth and verified actions for AI agents in authorized Chrome and Edge tabs.

Agent 控制浏览器：把 Chrome 变成本机 HTTP 控制端口，多 Agent 浏览器自动化，替代 MCP Chrome。Agent-controlled browser debugging: expose Chrome as a local HTTP control port for browser automation.

An MCP server that gives AI agents rich, programmatic control over Zen Browser, with first-class support for Zen-specific features such as Workspaces.

Control real Chrome from AI agents through a local daemon, reusable skill, and Manifest V3 extension.

A minimal local browser-control CLI and agent skill that drives one owned browser tab through a WebExtension bridge.

For those whom live chrome profile can't connect with LLM

南方星火 SouthernSpark — Open-source AI agent skills & browser automation tools. Build, share & monetize practical AI workflows. 开源 AI Agent 技能包 & 浏览器自动化工具箱。

Human-like browser control for AI agents — verify-first (proves every action worked), with an MCP server + JS & Python SDKs. Playwright + native + extension. MIT.

Ferris​ is an agent-like browser application (similar to OpenClaw) that can autonomously control web pages and search for information. It grants AI the permission to operate the browser, enabling automated navigation, form filling, data extraction, and other complex web tasks—all without human intervention.