Agentes e multiagente

BrowserOS: The Open-Source Agentic Browser

browseros-ai/browseros

BrowserOS is a free, open-source Chromium fork with a built-in AI agent, accompanied by BrowserOS neo, a secondary browser designed specifically for AI agents. It enables local-first browser automation, supporting multiple LLM providers and MCP-compatible tools.

★ 12,9KEstrelas
⑂ 1,4KForks
60Problemas em aberto
TypeScriptIdioma
AGPL-3.0Licença
Q@project.QualityScorePontuação editorial

Capturas de tela do projeto

Captura de tela de BrowserOS: The Open-Source Agentic Browser Captura de tela de BrowserOS: The Open-Source Agentic Browser Captura de tela de BrowserOS: The Open-Source Agentic Browser Captura de tela de BrowserOS: The Open-Source Agentic Browser Captura de tela de BrowserOS: The Open-Source Agentic Browser Captura de tela de BrowserOS: The Open-Source Agentic Browser

Visão geral

BrowserOS consists of two products maintained in a single monorepo. The primary BrowserOS browser integrates an AI agent directly into Chromium, allowing users to perform tasks via natural language with 20+ built-in tools and 40+ app integrations. BrowserOS neo is a dedicated browser for AI agents, allowing tools like Claude Code or Codex to run parallel web tasks using the user's imported logins. Both browsers run locally on the user's machine, preserving privacy and supporting Chrome extensions via one-click import.

Principais recursos

  • Two browsers: BrowserOS for human use with built-in AI, BrowserOS neo for AI agents
  • One-click import of Chrome logins, bookmarks, and extensions
  • Live dashboard to monitor parallel agent sessions
  • Session replay with scrubbable video and action timeline
  • Supports 11+ AI providers or fully local models via Ollama and LM Studio
  • MCP Server capability for integration with tools like Claude Code and Cursor
  • Local-only, privacy-first architecture storing data in ~/.browserclaw/
  • Real ad blocking with uBlock Origin (Manifest V2 support)
  • Cowork feature combining browser automation with local file operations

Requisitos, instalação e início rápido

Users can download pre-built binaries for BrowserOS neo (macOS, Windows) and BrowserOS (macOS, Windows, Linux AppImage, Debian) via the official website or docs. For developers, the agent monorepo can be set up following the packages/browseros-agent README. Building the Chromium fork requires C++/Python and ~100GB disk space, with instructions located in the packages/browseros directory.

Uso

Install the preferred browser (BrowserOS or BrowserOS neo). Import your Chrome profile with one click. For BrowserOS, connect your AI provider keys or configure local models via Ollama/LM Studio. For BrowserOS neo, connect your AI agents (Claude Code, Codex, etc.) via one-click MCP integration. Assign tasks via natural language and monitor progress via the neo dashboard or BrowserOS new tab.

Compatibilidade do modelo e casos de uso

BrowserOS supports 11+ AI providers, including Claude, OpenAI, Gemini, and ChatGPT Pro via OAuth, as well as local models via Ollama and LM Studio. BrowserOS neo is compatible with any AI tool that speaks MCP, including Claude Code, Codex, Cursor, VS Code, Zed, OpenCode, Hermes, OpenClaw, and Antigravity.

Observações sobre licença e riscos

Licensed under the AGPL-3.0 license. Copyright 2026 Felafax, Inc.

Editorial verification 2026-08-09: repository URL, owner, description, license and repository statistics were reviewed. License metadata: AGPL-3.0. README was fetched for the channel draft; re-check repository dependencies, releases and model terms before production use.

Lançamento e manutenção

Not stated in the repository metadata.

Firecrawl

firecrawl/firecrawl

★ 161,1KTypeScript

LangChain

langchain-ai/langchain

★ 143,6KPython

RAGFlow

infiniflow/ragflow

★ 86,7KGo