正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Your agents should use a remote browser built specifically for them, We revamped our Browser Skill to use our custom Stagehand CLI, It's faster than MCP tools and each run uses a new browser context, avoiding prompt injection. npx skills add browserbase/skills —skill browser

Browserbase

21,340 subscribers

41,144 次观看 • 6 个月前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

19 条评论

Browserbase 的头像

Browserbase6 个月前

Checkout our skill here:

Jake 的头像

Jake6 个月前

How does Stagehand CLI compare to Playwright directly? I've been using Playwright for agentic browser automation.

AKR 的头像

AKR6 个月前

We need a token usage comparison to other popular skills.

Mesh | Crypto x AI 的头像

Mesh | Crypto x AI6 个月前

Strong take. Fresh browser context per run is huge for prompt-injection resistance. The part most teams still miss: state recovery, permissions, retries, and audit trails. That’s what makes agent browsing production-ready.

Mimu | AI Tools & News 的头像

Mimu | AI Tools & News6 个月前

Stagehand CLI + fresh browser per run = zero prompt injection? This is the secure agent browser we’ve needed! Building workflows daily .... isolation is a game-changer. Would test it live and explore collab

yanno 的头像

yanno6 个月前

The new browser context per run is the underrated feature here. Prompt injection via browser state is a real attack surface most people ignore until it bites them. Local browser tools don't reset between agent runs — that's a meaningful difference for production workloads.

Kyle Jeong 的头像

Kyle Jeong6 个月前

it's 2026 your agents should use browsers purpose-built for them

Paul Klein IV 的头像

Paul Klein IV6 个月前

I am feeling skilled!!!

Stagehand 的头像

Stagehand6 个月前

built on stagehand so you know it's good🤘

Cody McLain 的头像

Cody McLain6 个月前

fresh browser contexts is huge. we had so many weird state issues with our automation until we started spinning up clean instances for each run

Shrey Pandya 的头像

Shrey Pandya6 个月前

I am speed

michael s galpert 的头像

michael s galpert6 个月前

🔥

Jackie.W 的头像

Jackie.W6 个月前

Can I login and let agent to use my login status to operate on browser pages?

Connor Grant 的头像

Connor Grant6 个月前

I've gotta move most of my preliminary prospect research to claude code & @browserbase, regular scraping isn't digging nearly as deep as this would 👀

Marcos Placona 的头像

Marcos Placona6 个月前

@grok can this be used to simulate a browser and cookies on an environment without a user interface? Like an Ubuntu server? Specifically can this be used to do things like create and edit Google Docs?

Agentic Development 的头像

Agentic Development6 个月前

@HeyGenLabs summarize

Ziray Hao 的头像

Ziray Hao6 个月前

another CLI gear drop for my agents

Dylan Normandin 的头像

Dylan Normandin6 个月前

does this help us get around bot detection algos?

RATIO. The Digital Court Society 的头像

RATIO. The Digital Court Society6 个月前

So does this mean i can use my mobile claude app and it will have access to a web browser? @grok

相关视频

Moltbot can now use Browser Use Skills! Here's Moltbot predicting the future with Polymarket and Kalshi skills 🔮 Browser Use Skills reverse engineer APIs to turn 30s browser workflows into 2s tool calls. We gave Moltbot: > Access to the Browser Use Skill Marketplace > Skill cloning and execution Try it now!

Moltbot can now use Browser Use Skills! Here's Moltbot predicting the future with Polymarket and Kalshi skills 🔮 Browser Use Skills reverse engineer APIs to turn 30s browser workflows into 2s tool calls. We gave Moltbot: > Access to the Browser Use Skill Marketplace > Skill cloning and execution Try it now!

Browser Use

35,382 次观看 • 7 个月前

We just built AWS Lambda with a browser built-in. Introducing Functions, a new compute paradigm that co-locates your code and browsers to build faster browser agents. Functions lets your team focus on the product, while we handle the infrastructure.

We just built AWS Lambda with a browser built-in. Introducing Functions, a new compute paradigm that co-locates your code and browsers to build faster browser agents. Functions lets your team focus on the product, while we handle the infrastructure.

Browserbase

63,284 次观看 • 7 个月前

We open-sourced our agent skills ecosystem, built specifically for complex end-to-end browser tasks. Type in any domain, and researches an approach to reliably complete your task, contributing it back to the knowledge base for other agents. Watch our agent fetch Amazon listings in real-time, driven entirely by a custom browser skill:

We open-sourced our agent skills ecosystem, built specifically for complex end-to-end browser tasks. Type in any domain, and researches an approach to reliably complete your task, contributing it back to the knowledge base for other agents. Watch our agent fetch Amazon listings in real-time, driven entirely by a custom browser skill:

Shrey Pandya

10,312 次观看 • 4 个月前

Introducing: Browser Use Skills 🛠️ [beta] Browser Agents are slow, non-deterministic, and expensive. Skills fix that. 🤯 You describe the workflow. We analyse the HTTP requests, reverse engineer the APIs, and turn them into tools calls your agents can use. ✅ 30s → 0.5s ✅ deterministic ✅ extremely cheap Want to run this at scale with Browser Use? DM me. Try it out for free 🔗↓

Introducing: Browser Use Skills 🛠️ [beta] Browser Agents are slow, non-deterministic, and expensive. Skills fix that. 🤯 You describe the workflow. We analyse the HTTP requests, reverse engineer the APIs, and turn them into tools calls your agents can use. ✅ 30s → 0.5s ✅ deterministic ✅ extremely cheap Want to run this at scale with Browser Use? DM me. Try it out for free 🔗↓

Gregor Zunic

60,548 次观看 • 9 个月前

there's no excuse to be unemployed, even browser agents are getting jobs with Stagehand 🤘 & Browserbase, you can spin up concurrent browser sessions to apply to infinite jobs get yourself (and/or your browser agents) a job with our templates

there's no excuse to be unemployed, even browser agents are getting jobs with Stagehand 🤘 & Browserbase, you can spin up concurrent browser sessions to apply to infinite jobs get yourself (and/or your browser agents) a job with our templates

Kyle Jeong

61,980 次观看 • 9 个月前

I built /schedule, but with access to your authenticated accounts in a secure browser. Using the Browserbase /cookie-sync skill, I saved my local browser cookies, and injected them into the remote browser in our schedulable Functions. The opportunities are endless, build Crons with CC to do ANYTHING a human can.

I built /schedule, but with access to your authenticated accounts in a secure browser. Using the Browserbase /cookie-sync skill, I saved my local browser cookies, and injected them into the remote browser in our schedulable Functions. The opportunities are endless, build Crons with CC to do ANYTHING a human can.

Kyle Jeong

49,399 次观看 • 5 个月前

Coding agents suck at using a browser. Playwright MCP burns through your context window before you even send your first prompt. That is why I built Dev Browser, a Claude Skill to let your agent close the loop without eating up tokens.

Coding agents suck at using a browser. Playwright MCP burns through your context window before you even send your first prompt. That is why I built Dev Browser, a Claude Skill to let your agent close the loop without eating up tokens.

Sawyer Hood

312,803 次观看 • 9 个月前

We're excited to announce our partnership with Prime Intellect to allow anyone to train browser agents. General-purpose models aren't optimized for your browser workflows, BrowserEnv lets you train one that is. Checkout and train your own custom model in a few hours.

We're excited to announce our partnership with Prime Intellect to allow anyone to train browser agents. General-purpose models aren't optimized for your browser workflows, BrowserEnv lets you train one that is. Checkout and train your own custom model in a few hours.

Browserbase

191,891 次观看 • 5 个月前

OpenClaw is all the rage and skills are a huge reason why. We've revamped ours to make it easier for Claude to go from local to prod. Check out how you can use Claude, Skills, and GitHub Actions to automatically self-heal browser automation scripts.

OpenClaw is all the rage and skills are a huge reason why. We've revamped ours to make it easier for Claude to go from local to prod. Check out how you can use Claude, Skills, and GitHub Actions to automatically self-heal browser automation scripts.

Browserbase

27,910 次观看 • 7 个月前

Introducing the new dev-browser cli. The fastest way for an agent to use a browser is to let it write code. Just `npm i -g dev-browser` and tell your agent to "use dev-browser"

Introducing the new dev-browser cli. The fastest way for an agent to use a browser is to let it write code. Just `npm i -g dev-browser` and tell your agent to "use dev-browser"

Sawyer Hood

870,294 次观看 • 5 个月前

Introducing the Search API, powered by Exa. In our benchmarking, browser agents are up to 3x faster when using integrated search. Your agents deserve the best search. Starting today, agents built on Browserbase get 1,000 free searches per month.

Introducing the Search API, powered by Exa. In our benchmarking, browser agents are up to 3x faster when using integrated search. Your agents deserve the best search. Starting today, agents built on Browserbase get 1,000 free searches per month.

Browserbase

68,776 次观看 • 6 个月前

Loved the skills system of the @Diabrowser and dropped a Skill Gallery for Dia Browser—one-click commands for content, research, and design, all in your browser. Link below.

Loved the skills system of the @Diabrowser and dropped a Skill Gallery for Dia Browser—one-click commands for content, research, and design, all in your browser. Link below.

Caner Aras

85,041 次观看 • 1 年前

We just solved authentication for AI Agents. Announcing 1Password + Browserbase, enabling secure agentic password autofill for your browser agents. Available exclusively on Director dot ai and Browserbase. Full post below.

We just solved authentication for AI Agents. Announcing 1Password + Browserbase, enabling secure agentic password autofill for your browser agents. Available exclusively on Director dot ai and Browserbase. Full post below.

Browserbase

185,095 次观看 • 11 个月前

Hermes has twelve browser tools. Browser Use mode replaces them with a single one, driven by Browser Use's CLI 3.0. Instead of a dozen schemas in every request and a tool call per click, the agent writes a script. In our tests that cut token use 48-66% with no accuracy drop.

Hermes has twelve browser tools. Browser Use mode replaces them with a single one, driven by Browser Use's CLI 3.0. Instead of a dozen schemas in every request and a tool call per click, the agent writes a script. In our tests that cut token use 48-66% with no accuracy drop.

Nous Research

837,764 次观看 • 1 个月前

Someone built a browser designed from the ground up for humans and AI agents to share. It's called ego lite. -Your agents run tasks in their own Spaces while your tabs stay yours -No extra setup - agents access your real logins and tabs directly through ego-browser -⁠Faster task completion on fewer tokens than any existing browser automation framework Every other tool makes you run a separate browser. This one just shares yours. Github:

Someone built a browser designed from the ground up for humans and AI agents to share. It's called ego lite. -Your agents run tasks in their own Spaces while your tabs stay yours -No extra setup - agents access your real logins and tabs directly through ego-browser -⁠Faster task completion on fewer tokens than any existing browser automation framework Every other tool makes you run a separate browser. This one just shares yours. Github:

0xMarioNawfal

51,768 次观看 • 1 个月前

Computer use is finally reliable. It just needed a real harness. Now there is one. Starting today, anyone can ship a SOTA browser agent, batteries included. Introducing Browserbase Agents: one prompt and one API call is all you need to automate the whole web.

Computer use is finally reliable. It just needed a real harness. Now there is one. Starting today, anyone can ship a SOTA browser agent, batteries included. Introducing Browserbase Agents: one prompt and one API call is all you need to automate the whole web.

Browserbase

96,219 次观看 • 2 个月前

UI Skills are now easier for agents. Introducing UI Skills CLI. ask Codex, Cursor, Claude Code, or any agent to use the CLI, like: "use `npx ui-skills start` to fix the motion in this dialog" it finds the best design-engineering skill for the job and uses it as context

UI Skills are now easier for agents. Introducing UI Skills CLI. ask Codex, Cursor, Claude Code, or any agent to use the CLI, like: "use `npx ui-skills start` to fix the motion in this dialog" it finds the best design-engineering skill for the job and uses it as context

Ibelick

62,979 次观看 • 3 个月前

Give Claude Code, Codex, OpenClaw, Cline access to the entire web from your terminal. Introducing TinyFish CLI + Skills. 15+ agent workflows ready to use. CLI runs browser automations from bash for max token efficiency. Skills tell agents how to use that CLI. 500 free steps.

Give Claude Code, Codex, OpenClaw, Cline access to the entire web from your terminal. Introducing TinyFish CLI + Skills. 15+ agent workflows ready to use. CLI runs browser automations from bash for max token efficiency. Skills tell agents how to use that CLI. 500 free steps.

TinyFish

183,436 次观看 • 5 个月前