#2121st of 24 in Agent frameworks and SDKs
stagehand
SDK for driving browsers with natural-language act, observe and extract calls
Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 25.6k
- License
- MIT
- Last commit
- Oct 2026
- Last release
- Aug 2026
- Language
- TypeScript
Missed the latest GitHub check; these facts may be older.
Overview
Stagehand is a browser automation SDK for TypeScript, Python and Go with Playwright-style page and locator methods. It adds LLM-driven act(), observe() and extract() calls, with schema-validated extraction and selectors returned so credentials stay out of the model. It runs against local Chrome or Browserbase's hosted browsers, where a model gateway and server-side caching are also available.
Who it is for: Developers building browser automations and web agents in TS, Python or Go
Strengths
- Same browser driver API across TypeScript, Python and Go SDKs
- observe() returns real selectors, so credentials can be filled without the model seeing them
- extract() validates output against Zod or Pydantic schemas, or Go types
- Deep locators for nested iframes and closed Shadow DOM; OTel traces
Weaknesses
- Local runs require Chrome installed
- Model gateway, server-side caching, proxies and session replay need a Browserbase account
- Local use needs an LLM provider API key, and each action costs tokens
- Fetch and Search add-ons use the Browserbase API key
What it needs
- no GPU
- Needs Chrome, LLM provider API key (e.g. OpenAI), Browserbase API key (optional, for cloud)
- Models: OpenAI (openai/gpt-5.4-mini in examples), Browserbase Model Gateway
Also in Agent frameworks and SDKs
See all 24| Rank | Project | Score |
|---|---|---|
| 1 | ppt-masterAgent skill that turns documents into natively editable PowerPoint decks | 91 out of 100 |
| 2 | langchainPython framework for building LLM agents and apps from interoperable components | 88 out of 100 |
| 3 | browser-usePython library that lets an LLM agent control a web browser | 88 out of 100 |
| #4 | tradingagentsLangGraph multi-agent framework simulating a trading firm to analyze tickers | 87 out of 100 |
| #5 | crewaiPython framework for multi-agent workflows with role-based Crews and event-driven Flows | 81 out of 100 |