Stagehand

the SDK for browser agents, built by @browserbase

join the discord:

[Stagehand @Stagehanddev]

Introducing our Browser Agent Evals.

We benchmark frontier and open models using various harnesses on computer-use tasks.

Our evals are fully open source and reproducible via our Evals CLI.

[Stagehand @Stagehanddev]

https://t.co/eBwgT9Z60K

[Browserbase @browserbase]

To do real work on your behalf, agents need to sign-in securely.

We’re excited to partner with 1Password so your agents can log in without ever seeing your passwords.

Agents request a login, you approve it in 1Password, then credentials are sent directly to the browser.

[Stagehand @Stagehanddev]

Read the full post:
https://t.co/NkZrpOarzG

[Stagehand @Stagehanddev]

You can extract typed data from any website using @pydantic and Stagehand.

Using Pydantic's enforced typing and Stagehand's structured extract, you can build robust and self-healing web scraping pipelines.

Read our full blog post below.

[Stagehand @Stagehanddev]

Full repo & examples here:

https://t.co/Z64FDpVfvE

[Stagehand @Stagehanddev]

Power your browser agents with Stagehand v4 in OpenAI's official SDKs:

https://t.co/wQLV8kLlcW

[Kevin Whinnery @kevinwhinnery]

Browser and computer use got waaaay easier in the latest @OpenAIDevs Python and Node.js SDKs 🧵

[Shrey Pandya @shreypandya]

New ship on https://t.co/cER97mCe5x - record mode!

-> fire up a browser session
-> show your agent what to do, click-by-click
-> hit export to download your finished workflow

you get a SKILL.md your agent can follow, plus scripts for @Stagehanddev, Playwright, or Puppeteer.

[Browserbase @browserbase]

Browserbase Search is now powered by Parallel (@p0) as the default search provider.

Parallel Fast returns sub-second results when speed matters. Parallel Advanced goes deeper for multi-step research when quality matters most.

[Stagehand @Stagehanddev]

See our benchmarks: https://t.co/inVUMLgsUb

[Stagehand @Stagehanddev]

Gemini 4 Argon is live in Stagehand.

It outperforms Fable 5.1 on computer use tasks with @eve and @pidotdev harnesses, while being roughly around the same cost/speed.

You can try the model in our playground at https://t.co/7bFLxRjpBE.

[Browserbase @browserbase]

We're sponsoring the AI Assistant meetup on Monday October 19th @ 7:30pm.

Come meet other builders and hear from teams building the infra powering the fastest growing assistants on the market.

Link to RSVP below.

[Shrey Pandya @shreypandya]

Coolest new feature in Stagehand - we can now automatically detect & invoke WebMCP tools on the fly.

This unlocks interacting with all @stripe checkout pages without repeatedly polling for new tools.

Give it a try on 4.2!

[Stagehand @Stagehanddev]

https://t.co/oFliUy1zh4

[Stagehand @Stagehanddev]

Find out how well your favorite models can use a computer:
https://t.co/inVUMLgsUb

[Stagehand @Stagehanddev]

Claude 5.5 Haiku is on par with GPT 6 Luna from OpenAI on accuracy and speed, but is 3x the cost per task.

As computer use improves, most gains will show up on the Pareto frontier: maintaining high accuracy while also being faster and cheaper.

[Charly Poly @whereischarly]

https://t.co/G7Thp5M7DZ

[Stagehand @Stagehanddev]

@ClaudeDevs https://t.co/C4busq5PVF

[Stagehand @Stagehanddev]

Stagehand supports @ClaudeDevs' new Tool SDK, converting model actions into browser actions.

Follow our quickstart guide to build your first browser agent with Claude and Stagehand.

https://t.co/RHX30QE6zA

[Stagehand @Stagehanddev]

Here are our docs to get started: https://t.co/lBIOOrFwCs

[Stagehand @Stagehanddev]

Stagehand supports @ClaudeDevs' new Tool SDK, converting model actions into browser actions.

Follow our quickstart guide to build your first browser agent with Claude and Stagehand.

https://t.co/RHX30QE6zA

[ClaudeDevs @ClaudeDevs]

Computer use and browser use toolsets are now built into the Python and TypeScript SDKs for Claude.

The API tells you what Claude wants to click or type.

Previously, you had to write your own loop and map clicks and keystrokes to commands, but the SDKs now run the loop and send actions to drivers.

[Stagehand @Stagehanddev]

Computer use models are getting really fast.

Opus 5 (released in July '26) averages 678s per task while newer models like Opus 5.5 average 326s, ~2x speed up.

See the speed up for yourself across 72 model x harness combinations at https://t.co/5YJbn7eBdV