AI Agent Platforms / All platforms / Skyvern
Skyvern
Browser automation platform that uses LLMs and computer vision to run website workflows.
Skyvern automates browser-based workflows using large language models and computer vision instead of brittle XPath selectors. Teams use it to fill forms, extract data, and run multi-step tasks on websites, self-hosted or through its cloud platform.
| Category | Browser agents |
| License | AGPL-3.0 |
| Hosting | self-hosted; vendor cloud available |
| Repository | github.com/Skyvern-AI/skyvern |
| Website | www.skyvern.com/docs |
| Last verified | 2026-09-02 |
Documented features
- Interacts with page elements from natural language prompts via page.act()
- Extracts structured data with optional JSON schemas via page.extract()
- Validates page state with page.validate() and runs multi-step tasks with an agent
- Builds workflows with loops, conditional logic, file parsing, HTTP requests, and code blocks
- Handles two-factor authentication over TOTP, email, and SMS
- Integrates password managers including Bitwarden, 1Password, and LastPass
- Streams a live viewport for debugging runs
- Downloads files and stores them, and sends emails from workflows
- Connects to Zapier, Make.com, and N8N
Feature summaries come from the project's own repository and documentation and describe the software as documented on the verification date, not an evaluation.