#795 · Primary category: AI Agents & Automation

HyperAgent

agent ai automation chrome llm playwright web

AI Browser Automation

Project last updated:05/11/26

GitHub Stars

1.5K

Forks

199

Contributors

7

License

Other

Why we included this project

Anyone who has kept Playwright or Selenium suites alive knows the pain: a site moves a button, renames a CSS class, and your carefully written selectors stop matching. HyperAgent puts an LLM on top of Playwright so you can say what you want in plain language, like asking for flight routes from Miami to New Orleans, and let the model work out which elements to touch. It splits the work into quick single actions, such as clicks and form fills, and longer multi-step tasks that can lean on screenshots when the DOM alone is unclear. There's also structured data extraction with typed schemas, an MCP client for pushing results into tools like Google Sheets, and action caching that replays recorded workflows without spending LLM tokens. If your automation keeps breaking on minor UI changes and you'd rather describe the goal than hand-write every selector, this is a solid middle ground.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category