HomeProjectsArticlesPK BoardAbout
Explore Projects

Skyvern

Active
GitHub Python AGPL-3.0

Description

Skyvern is an agent platform for browser task automation, using page understanding and action planning to complete complex web workflows such as forms and back-office tasks.

Key Features

  • Vision LLM driven — Uses vision models to understand and interact with websites, no predefined XPath needed
  • Playwright-compatible SDK — Adds AI capabilities on top of Playwright for Python and TypeScript
  • No-code workflow builder — Visual interface for non-technical users to create automation flows
  • Layout-change resistant — Visual understanding instead of DOM selectors, adapts to website changes
  • Multi-step task execution — Supports complex multi-step web operation workflows
  • Enterprise features — Built-in anti-bot detection, proxy networks, and CAPTCHA solving

Use Cases

💡 Automate form filling and submission workflows
💡 Batch process repetitive back-office operations
💡 Execute unified automation workflows across websites
💡 Data collection and web information extraction

Strengths & Limitations

✅ Strengths

  • • Actively maintained, recent updates
  • • High community interest (22.9k stars)
  • • Established track record (2 years in production)
  • • Responsive to issues, low backlog

⚠️ Limitations

  • • Restrictive license (AGPL-3.0)

Tags

browser automation web python

Categories

🌐 Browser Agent

Quick Start

pip install "skyvern[all]"
skyvern quickstart

# Or Docker Compose
git clone https://github.com/skyvern-ai/skyvern.git && cd skyvern
cp .env.example .env
docker compose up -d
# Open http://localhost:8080
Visit GitHub Visit Website View Docs

Related Projects

ScaleCUA

1.1k · Python
Stale C

Open-sourced computer use agents that can operate on cross-platform environments including Windows, macOS, Ubuntu, and Android. ICLR 2026 Oral paper project.

browserpythonagent +1
  • · Cross-platform GUI operation: supports Windows, macOS, Ubuntu, Android, and Web environments
  • · Large-scale training data: operation trajectories across 6 operating systems and 3 task domains
  • · State-of-the-art performance: MMBench-GUI L1-Hard 94.4%, OSWorld-G 60.6%, WebArena-Lite-v2 47.4%
View on GitHub

PyWinAssistant

1.3k · Python
Stale C

The first open-source Artificial Narrow Intelligence generalist agent that fully operates GUIs using only natural language. Uses Visualization-of-Thought and Chain-of-Thought reasoning for spatial perception and HID simulation.

browseragentpython +2
  • · First open-source Computer-Using-Agent that operates GUIs through native Windows Accessibility APIs without OCR or pixel-level imaging
  • · Symbolic spatial mapping with hierarchical element tracking via OS-native parent/child relationships and coordinate systems
  • · Natural language to GUI automation: parse intent, map to UI elements, and generate adaptive workflows
View on GitHub

Page Agent

28.9k · TypeScript
Active A+

Page Agent is a JavaScript in-page GUI agent by Alibaba that controls web interfaces with natural language, enabling automated form filling, page navigation, and element interaction.

browseragenttypescript +2
  • · In-page JavaScript GUI agent — no browser extension, Python, or headless browser required
  • · Text-based DOM manipulation without screenshots or multimodal LLMs
  • · Bring your own LLM — works with any OpenAI-compatible API
View full details → View on GitHub

Related Articles

browser agentbrowser-usestagehand

Browser Agent Data Extraction and Form Filling: From LLM Vision to DOM Selectors

A systematic engineering guide to Browser Agent across two high-frequency tasks (structured data extraction and automated form filling). Compares three action localization strategies (vision LLM, DOM selector, hybrid coordinate) and trade-offs across four frameworks: Browser-Use, Stagehand, Playwright MCP, and Skyvern.

Read more →