经 AI Skill Hub 精选评估，rayobrowse 获评「推荐使用」。这款Agent工作流在功能完整性、社区活跃度和易用性方面表现出色，AI 评分 6.8 分，适合有一定技术背景的用户使用。
rayobrowse 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排，将复杂的多步骤任务拆解为清晰的自动化流程，实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成，适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
rayobrowse 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排，将复杂的多步骤任务拆解为清晰的自动化流程，实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成，适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
# 方式一：pip 安装（推荐）
pip install rayobrowse
# 方式二：虚拟环境安装（推荐生产环境）
python -m venv .venv
source .venv/bin/activate # Windows: .venv\Scripts\activate
pip install rayobrowse
# 方式三：从源码安装（获取最新功能）
git clone https://github.com/rayobyte-data/rayobrowse
cd rayobrowse
pip install -e .
# 验证安装
python -c "import rayobrowse; print('安装成功')"
# 命令行使用
rayobrowse --help
# 基本用法
rayobrowse input_file -o output_file
# Python 代码中调用
import rayobrowse
# 示例
result = rayobrowse.process("input")
print(result)
# rayobrowse 配置文件示例（config.yml） app: name: "rayobrowse" debug: false log_level: "INFO" # 运行时指定配置文件 rayobrowse --config config.yml # 或通过环境变量配置 export RAYOBROWSE_API_KEY="your-key" export RAYOBROWSE_OUTPUT_DIR="./output"
<p align="center"> <img src="assets/rayobrowse.png" alt="rayobrowse"> </p>
<p align="center"> <a href="https://pypi.org/project/rayobrowse/"><img src="https://img.shields.io/pypi/v/rayobrowse" alt="PyPI"></a> <a href="https://www.npmjs.com/package/rayobrowse"><img src="https://img.shields.io/npm/v/rayobrowse" alt="npm"></a> <a href="LICENSE"><img src="https://img.shields.io/badge/license-MIT-blue" alt="License"></a> <a href="https://github.com/rayobyte-data/rayobrowse"><img src="https://img.shields.io/github/last-commit/rayobyte-data/rayobrowse" alt="Last Commit"></a> <br> <a href="https://github.com/rayobyte-data/rayobrowse"><img src="https://img.shields.io/github/stars/rayobyte-data/rayobrowse" alt="Stars"></a> <a href="https://pypi.org/project/rayobrowse/"><img src="https://img.shields.io/pepy/dt/rayobrowse?label=pypi&logo=pypi&logoColor=white" alt="PyPI Downloads"></a> <a href="https://www.npmjs.com/package/rayobrowse"><img src="https://img.shields.io/npm/dt/rayobrowse?label=npm&logo=npm&logoColor=white" alt="npm Downloads"></a> </p>
<p align="center"> <a href="https://docs.rayobrowse.com">Docs</a> · <a href="#quick-start">Quick Start</a> · <a href="#what-rayobrowse-handles">Fingerprints</a> · <a href="#self-hosted-and-cloud">Self-Hosted and Cloud</a> </p>
rayobrowse is a Chromium-based browser built to look like a real user in server-side automation. It gives each session a coherent fingerprint across browser APIs, graphics, fonts, language, timezone, WebRTC, automation surfaces, and network-facing behavior.
It exposes standard CDP, so you can use it from Playwright, Puppeteer, Selenium, Scrapy, OpenClaw, or your own automation code without learning a new browser API.
Rayobyte uses rayobrowse at billion-page-per-month scale on major websites for its web scraping API. We have been in the web scraping industry for 11 years, and this browser is part of the infrastructure we rely on ourselves.
import httpx from playwright.sync_api import sync_playwright
r = httpx.get( "http://localhost:9222/connect", params={"os": "windows", "headless": "false", "vnc": "true"}, timeout=120, ) r.raise_for_status() cdp_url = r.text.strip() vnc_url = r.headers.get("x-vnc-url") or "http://localhost:6080/vnc.html"
with sync_playwright() as p: browser = p.chromium.connect_over_cdp(cdp_url) context = browser.contexts[0] if browser.contexts else browser.new_context() page = context.pages[0] if context.pages else context.new_page() page.goto("https://browserscan.net/") print(page.title()) print(f"To view your browser in VNC go to: {vnc_url}") input("Press Enter to close the browser...") browser.close() ```
The browser is cleaned up when the CDP session closes. For long-running sessions, reconnects, and explicit lifecycle control, use the SDK or the session management APIs.
For SDKs, session reuse, VNC, proxies, and integration guides, use docs.rayobrowse.com.
The hardest version of this problem is running a stealth browser reliably at scale on Linux servers without GPUs. That is where scraping teams actually run production workloads, so that is where we put our effort.
Docker lets rayobrowse ship the browser binary, patched runtime, fonts, graphics stack, fingerprint engine, daemon, and VNC support as one reproducible environment. You can run the same browser locally, in CI, on bare metal, or across a cloud fleet without rebuilding a fragile desktop browser setup on every machine.
Self-hosted rayobrowse is free and unlimited. Docker is the only required runtime dependency.
git clone https://github.com/rayobyte-data/rayobrowse.git
cd rayobrowse
docker compose up -d
curl http://localhost:9222/healthAI Skill Hub 为第三方内容聚合平台，本页面信息基于公开数据整理，不对工具功能和质量作任何法律背书。
建议在沙箱或测试环境中充分验证后，再部署至生产环境，并做好必要的安全评估。
✅ MIT 协议 — 最宽松的开源协议之一，可自由商用、修改、分发，仅需保留版权声明。
AI Skill Hub 点评：rayobrowse 的核心功能完整，质量良好。对于自动化工程师和运维人员来说，这是一个值得纳入个人工具库的选择。建议先在非生产环境试用，再逐步推广。
| 原始名称 | rayobrowse |
| 原始描述 | 开源AI工作流：Stealth Chromium browser for web scraping and AI agents.。⭐166 · Python |
| Topics | workflowai-agentsanti-bot-detectionanti-detectantidetect-browserbot-detectionpython |
| GitHub | https://github.com/rayobyte-data/rayobrowse |
| License | MIT |
| 语言 | Python |
收录时间：2026-05-22 · 更新时间：2026-05-22 · License：MIT · AI Skill Hub 不对第三方内容的准确性作法律背书。
选择 Agent 类型，复制安装指令后粘贴到对应客户端