Scrapybara logo

Scrapybara

Freemium
-(0 评价)
0 已收藏

Remote virtual desktop infrastructure and unified API for AI computer-use agents

网站：https://scrapybara.com/
收录时间：2026年7月13日
月访问量：
10.0K
社交与联系：
AI agentscomputer useremote desktopbrowser automationPython SDKTypeScript SDK
访问网站
Scrapybara
访问官方网站

工具信息

什么是 Scrapybara？

Scrapybara hosts remote desktop instances (Ubuntu, Windows, Chromium browser) that AI agents can control through a unified API and SDKs. It enables developers to automate free-form computing tasks such as browsing, code execution, scraping, and multi-step workflows while handling scaling, auth, and environments.

核心功能

  • Remote desktop instances

    Instantly spin up desktop instances with browser, code execution, and filesystem capabilities at massive scale.

  • Unified computer-use API

    One unified interface for computer use models supporting structured outputs, multi-turn conversations, and custom tools.

  • Python and TypeScript SDKs

    Install via pip install scrapybara or npm install scrapybara and start instances with simple client calls.

  • Instance types: Ubuntu, Browser, Windows

    Feature-rich Linux desktop, lightweight Chromium browser, and full Windows desktop options.

  • Speed, scale, and stability

    Start hundreds of instances in milliseconds with minimal latency; orchestration and scaling are handled automatically.

  • Interactive stream

    Monitor and yield control of instances in real time.

  • Authenticated access

    Save and load authenticated websites for agents.

  • Session persistence

    Pause and resume instances as needed.

  • Agent tools and Act SDK

    Built-in tools such as ComputerTool, BashTool, and EditTool plus Act SDK for multi-step agent execution.

使用场景

  • #1

    Deploy intelligent computer-use agents

    Connect computer-use agents to reliable desktop infrastructure to automate any free-form computing task at scale.

  • #2

    Code testing and PR bots

    CodeCapy analyzes code changes and executes hundreds of parallel end-to-end UI tests in Scrapybara Ubuntu instances.

  • #3

    Website scraping

    Scrape and process any website using agents with browser and desktop control (e.g., CopyCapy).

  • #4

    Deep research in parallel

    Run deep research scraping workflows across many parallel instances.

  • #5

    Interactive agent demos and games

    Examples include dungeon-crawler style agent experiences that explore and act in environments.

适用人群：Developers building AI computer-use agentsTeams automating browser, desktop, and free-form computing workflowsBuilders of multi-agent systems that need scalable remote desktops

定价方案

Free accounts start with 10 compute hours, 100 agent credits, and 5 concurrent instances. Basic plan $29/month (100 compute hours/mo, 500 agent credits/mo, 25 concurrent instances). Pro plan $99/month (500 compute hours/mo, 2500 agent credits/mo, 100 concurrent instances). Enterprise is custom with self-hosting, dedicated support, and 99.9% uptime SLA. Each Act SDK/playground step (1 assistant message with tool calls) consumes 1 agent credit; credit top-ups available at $0.04 each, or use your own provider API key.

替代工具

NotebookLM

NotebookLM

Google NotebookLM is an AI research tool that analyzes your personal documents to deliver accurate, cited insights and creative ideas—making complex information easier to understand and use.

🇺🇸9.60%
31M
数据分析AI Research Tool+2
Glean

Glean

Enterprise AI platform that unifies search, assistants, and autonomous agents to accelerate work.

🇺🇸52.4%
3.5M
数据分析Enterprise Search+10
GPTGO

GPTGO

Free web search meets real‑time ChatGPT answers.

🇺🇸7.36%
106K
数据分析search engine+11
YOU

YOU

AI‑ready web search & content extraction APIs

🇺🇸25.3%
1.1M
数据分析AI+11
Rayyan

Rayyan

AI-powered platform for systematic literature reviews, screening, deduplication, data extraction, and evidence synthesis.

🇺🇸19.7%
1.2M
数据分析systematic review+11
Duck.ai

Duck.ai

Privacy-focused AI chat feature from DuckDuckGo.

🇺🇸53.6%
690M
数据分析privacy-focused+6
M

Metaphor Systems

Real‑time web search API for AI agents

🇺🇸40.3%
757K
数据分析search API+7
Marqo

Marqo

AI‑native search, recommendation, and merchandising platform that trains a dedicated LLM on your product catalog.

🇺🇸26.9%
27K
数据分析AI search+11

常见问题

What is Scrapybara?
Scrapybara is remote virtual desktop infrastructure and a unified API for running AI computer-use agents that control browsers, filesystems, and desktops at scale.
What instance types are available?
Scrapybara offers feature-rich Ubuntu Linux desktops, lightweight Chromium browser instances, and full Windows desktops.
Which SDKs does Scrapybara provide?
Python (pip install scrapybara) and TypeScript (npm install scrapybara) SDKs are available for starting and controlling instances.
How does Scrapybara pricing work?
It uses a freemium model. Free accounts include 10 compute hours, 100 agent credits, and 5 concurrent instances. Basic is $29/month and Pro is $99/month with higher limits; Enterprise is custom. Each Act SDK step consumes 1 agent credit unless you use your own provider API key.
What computer-use models can I connect?
Scrapybara supports computer-use agents including OpenAI (CUA) and Claude Computer Use through its unified API.
Can I monitor or control running instances interactively?
Yes. An interactive stream lets you monitor and yield control of instances in real time, and sessions can be paused and resumed.

Scrapybara 联系方式与社交媒体

访问官方网站

评价

-

分享你对这个工具的使用体验...

工具信息

定价模式Freemium
分类数据分析
二级分类
AI agentscomputer useremote desktopbrowser automationPython SDKTypeScript SDK
语言英语
最后更新2026年7月16日
认领这个工具