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。

Introducing the /autobrowse skill (inspired by Andrej Karpathy's autoresearch harness) Ask your agent to perform any task on the web: it explores the page using the Browserbase CLI, learns what went wrong in previous attempts, and iterates until it converges on a reliable workflow. Once token usage is optimized,... show more

Shrey Pandya

1,872 subscribers

230,316 次观看 • 2 个月前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

Say hello to open source deep research for your favorite agent harness. Our AI-Q agent skill packages the work of building a research pipeline into a portable skill. Drop it into your harness, and the agent delegates a research task to a local or hosted AI-Q server and gets back a detailed report with citations. See it in Codex below 👇

Say hello to open source deep research for your favorite agent harness. Our AI-Q agent skill packages the work of building a research pipeline into a portable skill. Drop it into your harness, and the agent delegates a research task to a local or hosted AI-Q server and gets back a detailed report with citations. See it in Codex below 👇

NVIDIA AI

76,237 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

88,833 次观看 • 18 天前

Introducing the /browser-trace skill, Give your agent 100% observability into its browser: dump network requests, DOM content, screenshots, and CDP logs into a searchable filesystem. Great for reverse engineering, autoresearch loops, and monitoring the situation ™️.

Introducing the /browser-trace skill, Give your agent 100% observability into its browser: dump network requests, DOM content, screenshots, and CDP logs into a searchable filesystem. Great for reverse engineering, autoresearch loops, and monitoring the situation ™️.

derek

250,829 次观看 • 2 个月前

Smooth by Smooth is a real-time AI browser agent - and the fastest and most reliable one yet. It can autonomously perform any task on the web using mouse and keyboard. Infinitely scalable, production-ready, and secure, it's now available as a serverless API with instant browser spin-up and persistent sessions.

Smooth by Smooth is a real-time AI browser agent - and the fastest and most reliable one yet. It can autonomously perform any task on the web using mouse and keyboard. Infinitely scalable, production-ready, and secure, it's now available as a serverless API with instant browser spin-up and persistent sessions.

Y Combinator

33,958 次观看 • 8 个月前

Introducing /ui-test Give your agent a PR, and it'll test your feature in a real browser, generating an HTML report with UI fixes The planner agent generates adversarial test cases to break your app, assigns them to subagents, and evaluates the page using the Browserbase CLI We reviewed a merged PR on Cal.com and found a bunch of small UI nits:

Introducing /ui-test Give your agent a PR, and it'll test your feature in a real browser, generating an HTML report with UI fixes The planner agent generates adversarial test cases to break your app, assigns them to subagents, and evaluates the page using the Browserbase CLI We reviewed a merged PR on Cal.com and found a bunch of small UI nits:

Shrey Pandya

57,480 次观看 • 3 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

866,775 次观看 • 3 个月前

I've been using Claude Code completely wrong. I gave it a custom skill and Browser CLI tools and letting it do work for me. It can open pages, click buttons, fill in forms all from your authenticated browser. Just published it to the marketplace, install it in 2 commands.

I've been using Claude Code completely wrong. I gave it a custom skill and Browser CLI tools and letting it do work for me. It can open pages, click buttons, fill in forms all from your authenticated browser. Just published it to the marketplace, install it in 2 commands.

Paul Klein IV

229,523 次观看 • 8 个月前

happy friday. sneak peek at some new features i'm building for the bankr agent: > sandboxed filesystem (your agent gets its own secure file system in the browser. you control what it can access. no agent running loose on your actual computer) > skill uploads (plugin new capabilities so your agent learns new things) > cli download & access > secure environment variables for api keys > github integration (connect your repo for reads and writes directly from your agent) all of this runs in a secure browser environment. no desktop app. no messy configuration. no downloading agent harnesses from the terminal. just a sandbox you control. in the video: i download an audit skill, run it against a smart contract, and save the report straight to the filesystem. then i ask the agent to pull all my 2026 transactions and write a csv for my accountant.

happy friday. sneak peek at some new features i'm building for the bankr agent: > sandboxed filesystem (your agent gets its own secure file system in the browser. you control what it can access. no agent running loose on your actual computer) > skill uploads (plugin new capabilities so your agent learns new things) > cli download & access > secure environment variables for api keys > github integration (connect your repo for reads and writes directly from your agent) all of this runs in a secure browser environment. no desktop app. no messy configuration. no downloading agent harnesses from the terminal. just a sandbox you control. in the video: i download an audit skill, run it against a smart contract, and save the report straight to the filesystem. then i ask the agent to pull all my 2026 transactions and write a csv for my accountant.

deployer

45,638 次观看 • 3 个月前

Your AI agents can now learn entire skill trees from the web. Meet the new HyperSkill Give it a topic. It reads the docs and builds a skill tree your agent can navigate.Browse the graph. Download. Drop into your project. Open source. Powered by Hyperbrowser.

Your AI agents can now learn entire skill trees from the web. Meet the new HyperSkill Give it a topic. It reads the docs and builds a skill tree your agent can navigate.Browse the graph. Download. Drop into your project. Open source. Powered by Hyperbrowser.

Hyperbrowser

95,806 次观看 • 4 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 • 2 个月前

Your Plan SUCKS! 👀 Inspired by the legend Andrej Karpathy himself, I created a new skill: LLM Council. Create better plans, by committee. Supports: > Codex CLI > Gemini CLI > Claude Code > OpenCode Call the skill with your feature that you'd like to build. It will ask you some clarifying question, and then launch up to four parallel planning agents to create a detailed plan. Once the plans are in, all plans are anonymized and the "The Judge" will critique and choose the best plan *OR* the best parts from all of them. Finally, it will output a final-plan, which you can review and refine. I even created a nice UI for you to review and refine your plans. This Skill has been tested on Linux only, but should work on other platforms. Please report any bugs! Links in the comments.

Your Plan SUCKS! 👀 Inspired by the legend Andrej Karpathy himself, I created a new skill: LLM Council. Create better plans, by committee. Supports: > Codex CLI > Gemini CLI > Claude Code > OpenCode Call the skill with your feature that you'd like to build. It will ask you some clarifying question, and then launch up to four parallel planning agents to create a detailed plan. Once the plans are in, all plans are anonymized and the "The Judge" will critique and choose the best plan OR the best parts from all of them. Finally, it will output a final-plan, which you can review and refine. I even created a nice UI for you to review and refine your plans. This Skill has been tested on Linux only, but should work on other platforms. Please report any bugs! Links in the comments.

am.will

69,010 次观看 • 5 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

61,422 次观看 • 1 个月前

Introducing: Agent Messenger The easiest way to let your agents talk to other agents. Dro a CLI + Skill into your agent and let it discover & talk to agents from your friends, coworkers and even strangers ⚠️ Live on Product Hunt - LINK BELOW - Please consider upvoting ⚠️

Introducing: Agent Messenger The easiest way to let your agents talk to other agents. Dro a CLI + Skill into your agent and let it discover & talk to agents from your friends, coworkers and even strangers ⚠️ Live on Product Hunt - LINK BELOW - Please consider upvoting ⚠️

Patrick Tobler

19,868 次观看 • 2 个月前

this video is the CLEAREST explanation of how claude skills + AI agents work and how to use them most people set up an AI agent and wonder why it keeps disappointing them. the context window is everything context is what the model assembles before it takes any action. think of it like everything the agent needs to read before it does anything. the quality of what goes in determines the quality of what comes out. the models are genuinely really good right now. claude and gpt are exceptional. the variable is almost always the context you give them. 1. agent.md files are mostly unnecessary every single line you put in an agent.md file gets added to every single conversation you have with your agent. a 1000 line file is around 7000 tokens burning on every run. the model already knows to use react. it can read your codebase. save the agent.md for proprietary information specific to your company that the model genuinely cannot know on its own. 2. skills are the actual unlock a skill.md file works differently. what loads into context is only the name and description, around 50 tokens. the full instructions only appear when the agent recognizes it needs that skill. so instead of 7000 tokens on every run you have 50. and the agent stays sharp because the context window stays lean. the closer you get to filling the context window the worse the agent performs, same way you perform worse when someone dumps 10 things on you at once. 3. here is how to actually build a skill the right way most people identify a workflow and immediately try to write the skill. what you want to do instead is run the workflow by hand with the agent first. walk it through every single step. tell it what to check, what good looks like, what bad looks like. correct it in real time. once you have had a full successful run from start to finish, tell the agent to review everything it just did and write the skill itself. it writes a better skill than you will because it has the full context of what actually worked in practice not in theory. 4. recursively building skills is how you go from frustrated to reliable when the skill breaks, and it will break, ask the agent exactly why it failed. it will tell you specifically what went wrong. fix it together in that same conversation. then tell it to update the skill file so that failure mode never happens again. ross mike did this five times with his youtube report generator. it now pulls from eight different data sources and runs flawlessly every single time without him touching it. 5. sub agents are something you earn not something you set up on day one start with one agent. build one workflow. turn it into one skill. once that works add another. ross mike has five sub agents now covering marketing, business, personal and more. it took months to get there and every single one exists because a workflow proved it deserved to exist. the people who set up 15 sub agents on day one and wonder why nothing works skipped all the steps that make the thing actually run. 6. your workflow is the thing the model cannot get anywhere else the model has been trained on everything. it knows more than you about most things. what it does not have is your specific process, your taste, your way of doing things. that is what skills capture. that is what makes your agent actually useful versus a generic one. downloading someone else's skill means downloading their context onto your setup and it will not work the way you want it to because it was never built around how you work. this is the clearest explanation of how agents actually work i have heard. Micky runs this stuff every single day and the results show it. full episode is now live on The Startup Ideas Podcast (SIP) 🧃 where you get your pods people charge for this sorta stuff i give away the sauce for free i just want you to win watch

this video is the CLEAREST explanation of how claude skills + AI agents work and how to use them most people set up an AI agent and wonder why it keeps disappointing them. the context window is everything context is what the model assembles before it takes any action. think of it like everything the agent needs to read before it does anything. the quality of what goes in determines the quality of what comes out. the models are genuinely really good right now. claude and gpt are exceptional. the variable is almost always the context you give them. 1. agent.md files are mostly unnecessary every single line you put in an agent.md file gets added to every single conversation you have with your agent. a 1000 line file is around 7000 tokens burning on every run. the model already knows to use react. it can read your codebase. save the agent.md for proprietary information specific to your company that the model genuinely cannot know on its own. 2. skills are the actual unlock a skill.md file works differently. what loads into context is only the name and description, around 50 tokens. the full instructions only appear when the agent recognizes it needs that skill. so instead of 7000 tokens on every run you have 50. and the agent stays sharp because the context window stays lean. the closer you get to filling the context window the worse the agent performs, same way you perform worse when someone dumps 10 things on you at once. 3. here is how to actually build a skill the right way most people identify a workflow and immediately try to write the skill. what you want to do instead is run the workflow by hand with the agent first. walk it through every single step. tell it what to check, what good looks like, what bad looks like. correct it in real time. once you have had a full successful run from start to finish, tell the agent to review everything it just did and write the skill itself. it writes a better skill than you will because it has the full context of what actually worked in practice not in theory. 4. recursively building skills is how you go from frustrated to reliable when the skill breaks, and it will break, ask the agent exactly why it failed. it will tell you specifically what went wrong. fix it together in that same conversation. then tell it to update the skill file so that failure mode never happens again. ross mike did this five times with his youtube report generator. it now pulls from eight different data sources and runs flawlessly every single time without him touching it. 5. sub agents are something you earn not something you set up on day one start with one agent. build one workflow. turn it into one skill. once that works add another. ross mike has five sub agents now covering marketing, business, personal and more. it took months to get there and every single one exists because a workflow proved it deserved to exist. the people who set up 15 sub agents on day one and wonder why nothing works skipped all the steps that make the thing actually run. 6. your workflow is the thing the model cannot get anywhere else the model has been trained on everything. it knows more than you about most things. what it does not have is your specific process, your taste, your way of doing things. that is what skills capture. that is what makes your agent actually useful versus a generic one. downloading someone else's skill means downloading their context onto your setup and it will not work the way you want it to because it was never built around how you work. this is the clearest explanation of how agents actually work i have heard. Micky runs this stuff every single day and the results show it. full episode is now live on The Startup Ideas Podcast (SIP) 🧃 where you get your pods people charge for this sorta stuff i give away the sauce for free i just want you to win watch

GREG ISENBERG

192,483 次观看 • 3 个月前

Introducing AI Brainstore! It’s a proof-of-concept of a brain for an AI agent. Ask an agent a question, it checks its memories for an answer, otherwise it browses the web and learns the answer. As it learns, it saves memories to its brain. GitHub:

Introducing AI Brainstore! It’s a proof-of-concept of a brain for an AI agent. Ask an agent a question, it checks its memories for an answer, otherwise it browses the web and learns the answer. As it learns, it saves memories to its brain. GitHub:

Mckay Wrigley

514,642 次观看 • 3 年前

.EigenPal is AGI for backoffice paperwork: describe a process to automate, give it test data, and the AI agent iterates until the workflow is perfect. Try on Congrats on the launch Jedr Blaszyk & Matej Novak!

.EigenPal is AGI for backoffice paperwork: describe a process to automate, give it test data, and the AI agent iterates until the workflow is perfect. Try on Congrats on the launch Jedr Blaszyk & Matej Novak!

Y Combinator

21,829 次观看 • 5 个月前

Introducing /orchestrate, a skill that recursively spawns agents to tackle your most ambitious tasks with the Cursor SDK. We’ve used it to: - Autoresearch our internal skills, cutting token use by 20% while improving evals - Cut cold start times on our internal backend by 80%

Introducing /orchestrate, a skill that recursively spawns agents to tackle your most ambitious tasks with the Cursor SDK. We’ve used it to: - Autoresearch our internal skills, cutting token use by 20% while improving evals - Cut cold start times on our internal backend by 80%

Cursor

534,315 次观看 • 2 个月前

this is insane. just toggle this button and any coding agent can use your browser > no more Chrome extensions > One button, connect Claude Code to your browser all you need is the right harness... try it with the Browser Use CLI right now!

this is insane. just toggle this button and any coding agent can use your browser > no more Chrome extensions > One button, connect Claude Code to your browser all you need is the right harness... try it with the Browser Use CLI right now!

shawn

639,600 次观看 • 4 个月前

you've bookmarked 247 winning ads on twitter. you've shipped maybe 2. the gap was never strategy. it was the production cycle between bookmark and live. the workflow that closes it sounds fake until you try it once. here's the workflow. → paste the winning ad into claude. ask for the teardown. → claude returns the hook, angle, offer, and a ready-to-use prompt. → paste into notch ad agent. it doesn't generate. it strategizes and ships. on-brand version renders. → push to meta. same day. no designer. no notion. no waiting. while the teardown thread is still arguing why the ad won, your version is collecting data on meta. drop a winning ad link in the comments. we'll spin one up for you.

Vinay Jain

617,739 次观看 • 1 个月前

Everything is better in blocks. Introducing the new Voxel-It Skill—upload any photo and watch it turn into a Minecraft-style world. The Skill voxelizes the background, object, or both, while keeping the people perfectly human.

Everything is better in blocks. Introducing the new Voxel-It Skill—upload any photo and watch it turn into a Minecraft-style world. The Skill voxelizes the background, object, or both, while keeping the people perfectly human.

Pika

22,227 次观看 • 16 天前