Back to directory

Stagehand

An SDK for building browser agents that extract data from and interact with any website, using Playwright-style methods across TypeScript, Python and Go.

Stagehand

Stagehand is an SDK for browser agents that lets an agent sign in once, keep the session, and pull structured data out the other side. It uses Playwright-style methods such as goto, click, locator and screenshot that developers and agents already know, and gives agents self-healing act, observe and extract operations that adjust how an action happens when a site's layout changes.

It trims the accessibility tree to give an agent only the page context it needs, reducing token use, and runs as an extension alongside the browser to cut round-trip latency. It supports agent-oriented features such as WebMCP, clipboard access, batch commands, and deep locators for nested iframes and closed shadow DOMs, plus OpenTelemetry traces. The same driver is available in TypeScript, Python and Go.

Local runs need Chrome installed; it can also run on the Browserbase cloud service, which adds a model gateway, caching, residential proxies and session recordings. A hosted Browserbase MCP server exposes navigate, act, observe and extract to any MCP client without a local browser. It is licensed under the MIT license.

More in Browser automation

See all 88
  • Loads an unpacked Chrome extension that lets a local AI coding agent click, type and read the tabs, cookies and logins of a Chrome window you already have open.

  • Hronaut

    hronaut.dev

    A visible, persistent Electron browser that exposes durable, named agent workspaces over MCP, so a coding agent's browser session stays open and inspectable independently of any one AI conversation.

  • Prompt OS

    github.com

    A desktop AI agent for Windows that executes terminal commands, manages files, writes and runs code, browses the web and reads the screen, acting on natural-language requests rather than only answering them.