macOS
Homebrew Cask已验证 · 100%
brew install --cask browseros-ai/tap/browseros-neolocal Homebrew cask metadata
安装
其他安装命令
概览
软件包摘要
Browser for agents
历史
项目历史与用法
BrowserOS neo is a free, open-source secondary browser that lets MCP-compatible AI agents operate authenticated web sessions while users watch and replay their actions.
项目历史
BrowserOS originated in 2025 as an open-source Chromium-based agentic-browser project. The repository now ships two related products: BrowserOS, an AI browser driven by a person, and BrowserOS neo, a secondary browser driven by external AI agents.
Neo combines a Chromium fork with a Rust MCP and JSON backend plus a browser dashboard. Its product design centers on imported browser state, parallel agent sessions, live observation, and replayable local audit trails.
采用历史
BrowserOS neo targets users of MCP-capable agents including Claude Code, Codex, Cursor, VS Code, Zed, OpenClaw, and Hermes. Its one-click connection flow and Chrome import are designed to reduce the setup friction that commonly limits authenticated browser automation.
使用方式
Users install neo, import logins, bookmarks, extensions, and settings from Chrome, then connect an MCP-capable agent. Agent work appears in isolated tabs and in a new-tab dashboard, with each session available for later replay.
Official documentation says session history, screenshots, settings, and replay data remain on the local computer under ~/.browserclaw/. The browser is intended for authenticated tasks such as inbox work, filing expenses, updating web applications, and retrieving reports.
为什么软件包爱好者会关心
BrowserOS neo represents a shift from disposable headless-browser sessions toward a locally installed, inspectable browser environment for agents. Package users get a reproducible desktop installation while retaining real browser state, isolated agent tabs, an MCP endpoint, and on-disk audit artifacts.
时间线
- 2025: The BrowserOS project identifies its origin year in the repository's official citation metadata.
- 2026: BrowserOS neo is documented as a distinct agent-driven browser shipping from the BrowserOS monorepo.
Related projects
- BrowserOS
- Chromium
- Model Context Protocol
- Claude Code
- Codex
- Cursor
来源
安全态势
尚未找到受保护工具覆盖
没有找到 BrowserOS neo 的匹配本地密钥处理 manifest。软件包元数据仍在此发布，以便未来覆盖拥有稳定的软件包 URL。
安装行为
- 未记录 Homebrew bottle 元数据。
建议审查
在无人值守的代理使用前，请检查该工具是否读取明文凭据、写入远程状态、发布制品或调用插件。
local files
Configuration and credential file locations
These source-backed paths show where this package keeps local settings or durable credentials. Automic Vault can use them as review targets for secret scanning, migration, and command approval.
Configuration files
Config paths the tool may read or write during local use.
- Unix
~/.browserclaw/
可执行文件
已安装的可执行文件
| 命令 | 类型 | 暴露范围 | 备注 |
|---|---|---|---|
| 没有可执行文件数据。 | |||
新鲜度
版本和新鲜度
这些信号区分页生成时间、软件包管理器活动和上游发布比较。只有存在证据 URL 和可比较版本时，才会提示版本落后。
页面生成时间2026-10-22
管理器版本0.51.1.2
管理器更新时间
本地数据OK
上游not checked
检测到的最新版本未检测到
- 信息No package-manager update timestamp was available.低 置信度
- 信息Release/tag comparison is only available for GitHub repositories.https://browseros.com/none 置信度
安装元数据
软件包元数据
| 软件包键 | cask:browseros-ai/tap/browseros-neo |
|---|---|
| 版本 | 0.51.1.2 |
| 软件包管理器 | Homebrew Cask |
| 软件包管理器页面 | https://github.com/browseros-ai/homebrew-tap/blob/53539f3d68fbd92eeedcfcb53da062194d7ec8eb/Casks/browseros-neo.rb |
| 主页 | https://browseros.com/ |
| 上游文档 | https://browseros.com/ |
| 源码归档 | https://cdn.browseros.com/releases/browserclaw/0.51.1.2/macos/BrowserOS_neo_v0.51.1.2_universal.dmg |
| SHA-256 | 1e4755ee08573766098de6fb665b97cafd1917c58f41370c489b31f1a57ac408 |
| 下载 URL | https://cdn.browseros.com/releases/browserclaw/0.51.1.2/macos/BrowserOS_neo_v0.51.1.2_universal.dmg |
| Bottle | 未记录 |
| 服务 | 未声明 |
来源线索
由仓库数据生成
此页面由 av-web 从 scripts/generate-pkg-sqlite.py 生成的私有软件包 SQLite 工件提供。
Combined YAML source
View the package source record on GitHub.
使用的来源
- cross-ecosystem install command graph
- curated configuration and credential file locations
- curated package history
- package relationship graph
- package version freshness
- pkg.so package database