Introducing Claude Managed Agents: everything you need to build and deploy agents at scale. It pairs an agent harness tuned for performance with prod...
Claude Managed Agents provide a performance-tuned agent harness integrated with production infrastructure, enabling rapid transition from prototype to launch within days. Currently in public beta, this tool streamlines the agent deployment process, making it accessible for developers looking to scale their applications.
New in Claude Code: /ultraplan Claude builds an implementation plan for you on the web. You can read it and edit it, then run the plan on the web or...
New in Claude Code: /ultraplan Claude builds an implementation plan for you on the web. You can read it and edit it, then run the plan on the web or back in your terminal. Available now in preview for all users with CC on the web enabled. https://t.co/44nG7XGC1a
I put the entire Claude Code GTM Engineering Playbook into ONE Notion doc. 8 sections. No fluff. - How to get set up correctly from day one: Pro pla...
This document consolidates the Claude Code GTM Engineering Playbook into a single Notion doc, eliminating unnecessary content. It includes essential setups such as Pro plan installation across different operating systems and GUI installation methods, aimed at helping new users start with the platform effectively.
Autonomous AI researchers are here. Try @thesis_labs https://t.co/mTYGWQ6NqV
Autonomous AI researchers are here. Try @thesis_labs https://t.co/mTYGWQ6NqV
claude code just shipped /ultraplan. the speed isn't the interesting part. in under 5 minutes you'll learn: 1. how 4 opus agents plan your project i...
Claude's Ultraplan features a project planning method utilizing four agents: three explorers and one critic. While it boasts a 4x speed improvement, the article emphasizes the significance of the review workflow, presenting five scenarios where Ultraplan enhances project outcomes.
Free and open-source AI-first knowledge base and startup OS. Markdown files on disk. AI agents that actually work. No database. No vendor lock-in. Sel...
Cabinet is an AI-first knowledge base that operates on markdown files without any database requirements, allowing for a fully self-hosted, open-source experience. It employs AI agents to help users manage and execute tasks related to project management, sales outreach, and content creation. By eliminating vendor lock-in, Cabinet provides a customizable solution for various workflows, from startups to B2C and B2B applications.
Claude now supports dynamic looping. If you run /loop without passing an interval, Claude will dynamically schedule the next tick based on your task....
Claude now supports dynamic looping. If you run /loop without passing an interval, Claude will dynamically schedule the next tick based on your task. It also may directly use the Monitor tool to bypass polling altogether /loop check CI on my PR https://t.co/L5UFH22r0C
What does every big company think about the agent harness? Anthropic, OpenAI, CrewAI, LangChain. They all build agents. They all wrap their models in...
This content examines the varying perspectives of major AI companies like Anthropic, OpenAI, CrewAI, and LangChain regarding the concept of an 'agent harness.' While there is consensus on the utility of building infrastructure around AI models, opinions diverge on almost every other aspect of implementation, philosophy, and approach. It's a critical look at how different minds conceptualize and adapt to AI technology.
OpenAI is working on a new experimental feature for Codex called Scratchpad. Users will be able to start multiple Codex chats from a TODO list view,...
OpenAI is developing an experimental feature named Scratchpad for Codex, allowing users to initiate multiple Codex chats from a TODO list view and execute them in parallel. This feature aims to enhance productivity within the upcoming Codex Superapp.
Generate a personal engineering style wiki from your GitHub commit history - arlington-labs/gitstyle
Gitstyle is a tool that creates a personalized engineering style wiki based on your GitHub commit history. It analyzes your coding patterns and style preferences, allowing you to document and reflect on your development practices. By transforming commits into a structured format, it enhances self-awareness in coding techniques and can serve as a useful reference for future projects.
https://t.co/85spxn8PXg
https://t.co/85spxn8PXg
Optimize prompts, code, and more with AI-powered Reflective Text Evolution - gepa-ai/gepa
The gepa AI project on GitHub focuses on enhancing prompt optimization and code generation through a method called Reflective Text Evolution. This involves a strategic approach to improving the quality and effectiveness of AI-generated outputs. The project showcases tools and frameworks beneficial for developers aiming to leverage AI in software development.
you'll need to explicitly prompt Claude Code to use it, but the Monitor Tool is super powerful e.g. "start my dev server and use the MonitorTool to o...
you'll need to explicitly prompt Claude Code to use it, but the Monitor Tool is super powerful e.g. "start my dev server and use the MonitorTool to observe for errors" https://t.co/xVUr8MysOi
On coding agents, malleable software, and the future of interface invention
The article critiques the traditional app model and programming specialization as limiting factors in software development and user creativity. It argues for the potential of coding agents to create more malleable and composable digital interfaces, enriching the user experience beyond the constraints of one-size-fits-all applications. The author highlights the need for diversity in interface invention, calling for contributions from those without programming expertise.
Karpathy's LLM Wiki got 5,000 stars in 48 hours. Now someone extended it with the features it was missing. Memory lifecycle. Confidence scoring. Know...
LLM Wiki v2 has been launched, extending the original AI-generated wiki concept. Key features of the update include memory lifecycle management, confidence scoring, knowledge graphs, automated hooks, and forgetting curves. The original LLM Wiki gained significant popularity, achieving 5,000 stars in just 48 hours. This revision aims to enhance the functionality and usability of the platform.
Aggressive prediction: the CEOs like @jack and @tobi who are slinging code and open source and all the way at the edge are leading from the front. The...
Aggressive prediction: the CEOs like @jack and @tobi who are slinging code and open source and all the way at the edge are leading from the front. Their companies will make it Others still in manager mode? Oof maybe less so https://t.co/9Kwd4VyEfu
what it looks like setting up your second brain left side is my LLM knowledge base (based on Karpathy's approach) right side is unprocessed raw file...
This content showcases a personal implementation of a 'second brain' using an LLM knowledge base, inspired by Andrej Karpathy’s methods. It contrasts processed knowledge on the left with raw files awaiting integration on the right, emphasizing a quick build-out of a personal knowledge system. A Git repository link is provided for others to replicate the setup.
New item in my SOUL md tonight https://t.co/iAwcQ2QCB7
New item in my SOUL md tonight https://t.co/iAwcQ2QCB7
https://t.co/EvpBN6owh5
https://t.co/EvpBN6owh5
for those of you who are autoresearch pilled , or have been meaning to get into autoresearch but dont know how - I shipped evo today - a opensource Cl...
Evo is an open-source Claude Code plugin designed for autoresearch enthusiasts. It automates the code optimization process by analyzing a codebase, establishing benchmarks, and running baseline experiments to improve performance. This tool is particularly valuable for those looking to enhance their coding practices through empirical experimentation.
I’m building a way to turn websites into semantic design docs AI agents can actually understand. Like Pinterest, but for agents. Already generated a...
I’m building a way to turn websites into semantic design docs AI agents can actually understand. Like Pinterest, but for agents. Already generated a few hundred DESIGN.md examples so agents can study how good design actually works. Launching soon. https://t.co/CdSUnjooql
Want to put something in CLAUDE.md? Stick it in CODE_STANDARDS.md instead. Then pass it to a reviewer agent that runs on every PR. Save tokens duri...
Want to put something in CLAUDE.md? Stick it in CODE_STANDARDS.md instead. Then pass it to a reviewer agent that runs on every PR. Save tokens during implementation, spend them during review.
https://t.co/oQ2GEK1iAy https://t.co/w0SOdCeBBG
https://t.co/oQ2GEK1iAy https://t.co/w0SOdCeBBG
Dogfooding Opus 4.7 the last few weeks, I've been feeling incredibly productive. Sharing a few tips to get more out of 4.7 🧵
Dogfooding Opus 4.7 the last few weeks, I've been feeling incredibly productive. Sharing a few tips to get more out of 4.7 🧵
2/ The new /fewer-permission-prompts skill We've also released a new /fewer-permission-prompts skill. It scans through your session history to find c...
This new skill scans session history to identify safe, commonly used bash and MCP commands that trigger excessive permission prompts. It provides recommendations for commands to add to avoid these repetitive interruptions, enhancing user experience and efficiency.
Introducing Big Terminal Mode! ⌘⇧T to open a full terminal in the center panel. Run Claude, Codex, OpenCode, Amp all from Conductor: https://t.co/99...
Introducing Big Terminal Mode! ⌘⇧T to open a full terminal in the center panel. Run Claude, Codex, OpenCode, Amp all from Conductor: https://t.co/99mzJYH4sx
https://t.co/Ljzw0lmvao
https://t.co/Ljzw0lmvao
Introducing Claude Design by Anthropic Labs: make prototypes, slides, and one-pagers by talking to Claude. Powered by Claude Opus 4.7, our most capab...
Claude Design allows users to create prototypes, slides, and one-pagers through conversational interactions with the Claude Opus 4.7 AI model. This tool is in research preview, available for Pro, Max, Team, and Enterprise plans, and is rolling out gradually. It showcases the growing capabilities of generative AI in design and collaboration.
Garry's Opinionated OpenClaw/Hermes Agent Brain. Contribute to garrytan/gbrain development by creating an account on GitHub.
This repository showcases Garry Tan's implementation of the OpenClaw/Hermes Agent Brain, which likely incorporates personal preferences and methodologies for developing intelligent agents. It serves as a base for experimentation with AI-driven applications, bridging concepts of agent design with practical coding examples.
Summarize or extract text/transcripts from URLs, podcasts, and local files (great fallback for “transcribe this YouTube/video”).. Score: /10. Best for...
The Claude skill functionality allows users to summarize or extract text from various sources including URLs, local files, and podcasts. It employs a decision tree to parse user input and apply the appropriate processing methods based on content type and user-defined flags. The skill integrates with various API services to enhance its capabilities, though potential issues might arise if the necessary API keys are missing.
Frank Shotwell of Crux Capacity demonstrates how to build a multi-agent knowledge graph using Claude Code and GitHub for orchestration, teaching atten...
This content presents a metaphorical exploration of a community-driven project in Los Angeles, described as building a 'Bamboo Cathedral.' It illustrates how diverse agents work together, merging various perspectives and creativity into a collective endeavor. The piece emphasizes the importance of collaboration and the fusion of ideas to foster innovation.
karpathy is showing one of the simplest AI architectures that actually works.. dump research into a folder, let the model organise it into a wiki, as...
Andrej Karpathy presents a straightforward AI architecture that organizes research into a self-improving knowledge base. By dumping information into a folder, the model can create a wiki-like structure and refine itself with each query, leading to compound improvements. This insight highlights the importance of iterative feedback in knowledge systems.
https://t.co/nKTDfC7zMm
https://t.co/nKTDfC7zMm
Introduction Today we’re announcing Project Glasswing 1, a new initiative that brings together Amazon Web Services, Anthropic, Apple, Broadcom, Cisco...
Project Glasswing is a coalition formed by major tech companies like AWS, Google, and Microsoft, leveraging Anthropic's advanced AI model, Claude Mythos 2 Preview, to enhance cybersecurity. The initiative aims to proactively address vulnerabilities in critical software infrastructures before they can be exploited by malicious actors. With significant investment and collaboration across numerous organizations, the project emphasizes the urgency of improving cybersecurity in the wake of advancing AI capabilities.
Made an updated version this weekend Here's how you do it (raw notes) > Grab @karpathy's latest gist (in the first comment) > Download @steipe...
This content outlines an updated workflow for personal knowledge management, integrating various tools and APIs. Key components include downloading the summarize CLI, yt-dlp, Obsidian, and a tool by @tobi. The setup involves creating a CLI called 'brain' to streamline the process.
Hands-free email triage for your commute.
Voicemail AI Hands-free email and calendar for your commute. Connect your Google account to get started. [Connect Google](https://voice-email-production.up.railway.app/api/auth)
Anthropic's agent harness has a 823-line retry system. Your agent has try/catch. Most agent teams will spend 6 months discovering what's already si...
This content provides insights into Anthropic's agent harness, particularly its intricate 823-line retry system and how it compares to traditional agent handling with try/catch. It notes that many agent teams will waste significant time uncovering existing functionalities within the Claude Code's 331 modules, making the author's detailed analysis a valuable resource for understanding this framework.
tldr: everyone is converging on the same product shape: a general harness that takes a goal, uses tools, and does knowledge work. once every product...
tldr: everyone is converging on the same product shape: a general harness that takes a goal, uses tools, and does knowledge work. once every product is a harness, the next frontier is the feedback loop that improves it after deployment. https://t.co/6iDmHWOGJM
GStack users! Here's a tip: if you need something to be super well reviewed, i do "/plan-eng-review with full claude and codex review, THEN full cla...
This content outlines a review strategy tailored for GStack users, emphasizing the use of '/plan-eng-review' to conduct comprehensive reviews with Claude and Codex. The approach is particularly effective for security and performance optimization tasks, ensuring thorough evaluation from both a constructive and adversarial perspective.
https://t.co/i9kNLnWsvY
https://t.co/i9kNLnWsvY
I built /schedule, but with access to your authenticated accounts in a secure browser. Using the @browserbase /cookie-sync skill, I saved my local b...
This project outlines the creation of a scheduling feature that integrates authenticated access to user accounts through a secure remote browser. By leveraging the @browserbase /cookie-sync skill, the author successfully synchronizes local browser cookies, enabling the remote browser to access authenticated sessions seamlessly. This opens up numerous possibilities for enhanced user interactions in applications.
The next era of product development is built on context and agency. Here's how Linear is evolving. And what we're launching today.
Linear argues that traditional issue tracking is outdated due to its reliance on cumbersome handoff processes between PMs and engineers. Instead, it advocates for a system that leverages coding agents to simplify planning and execution by understanding user intent and context. The introduction of features like Linear Agent and automation signifies a shift towards a more context-driven approach in software development.
Introducing /cookie-sync Run browser tasks in the cloud with all your authenticated accounts, powered by @browserbase Watch as my agent: - uploads m...
Cookie Sync allows users to run browser tasks in the cloud while maintaining authenticated accounts by uploading local Chrome cookies to a remote browser. This approach automates personal tasks, exemplified by the demonstration of choosing a lunch order through Forkable.
Introducing Hyperagents: an AI system that not only improves at solving tasks, but also improves how it improves itself. The Darwin Gödel Machine (DG...
The Hyperagents system leverages concepts from the Darwin Gödel Machine (DGM) to facilitate AI's self-improvement capabilities both in task performance and in enhancing its own improvement processes. By iteratively generating and assessing better agents, it promises a new frontier in open-ended AI development.
https://t.co/I2Hf40EpIt
https://t.co/I2Hf40EpIt
This looks interesting https://t.co/18g8ym19rQ
This looks interesting https://t.co/18g8ym19rQ
Build interactive experiences, websites, 3d, and apps using natural language.
Omma is a platform for creating voxel-based websites, allowing users to remix and vary colors and components. The tool emphasizes community collaboration and ease of access by offering various components and chat functionalities for shared building experiences. The platform appears to cater to both individuals and teams looking to innovate in web design using 3D voxel elements.
Claude Code can now auto-fix your PR in the background! all you have to do is turn on the Auto Fix setting and go touch grass 🙏 https://t.co/TzUJ7A...
Claude Code can now auto-fix your PR in the background! all you have to do is turn on the Auto Fix setting and go touch grass 🙏 https://t.co/TzUJ7AomYd https://t.co/YIT0j3DgPS
This is actually insanely useful and amazing to see available OSS. We've also built this into @DevinAI and is available for all users: 1. Kick off ag...
This open-source tool provides automated testing and code completion functionalities, proving to be a valuable resource for developers. It integrates seamlessly with DevinAI, allowing users to initiate agent sessions, conduct automated tests, and obtain video recordings of app performance. This tool enhances development efficiency through programmatic testing and immediate code suggestions.
I guess gstack is now in distribution in Claude Code You can just open a blank window and say install gstack and it works now https://t.co/w4ERSXrjFC
I guess gstack is now in distribution in Claude Code You can just open a blank window and say install gstack and it works now https://t.co/w4ERSXrjFC
By default, Claude Code asks users for approval before running commands or modifying files. This keeps users safe, but it also means a lot of clicking...
Claude Code introduces an 'auto mode' to balance user safety and efficiency by automating permission approvals. This mode, leveraging model-based classifiers, aims to minimize approval fatigue while still preventing dangerous actions through input and output defense layers. The solution reflects a middle-ground approach between full manual control and unrestricted operation.
excited to share a tiny project I've been working on over the past few months called whoami wiki, it is a personal encyclopedia written by agents! it...
Whoami Wiki is a personal encyclopedia project designed to transform your digital archives, including photos, chats, and documents, into an interactive and living resource. This tool aims to help users document and reflect on their experiences and history through a curated digital narrative.
OMG you guys, this is incredible! This is using Google's new WebMCP function to control your browser, but not only is it lightning fast, but its uniqu...
The new WebMCP function in Google Chrome offers an impressive speed advantage by operating directly within the main browser instance. This approach contrasts with traditional automation tools like Playwright, which typically use sandboxed environments that may have limitations such as memory retention.
this guy built an app where AI agents autonomously create tasks, review each other's work, and message each other you just sit there and watch everyt...
This content describes an innovative application where AI agents autonomously handle task management by creating, assigning, and reviewing each other's work. Users can observe this real-time interaction on a dashboard, highlighting a novel approach to task automation and AI collaboration.
idk how ppl always find new usecases for the stuff we make, this is an entire graph made for studying a biology paper https://t.co/pmwsiYJJ01
idk how ppl always find new usecases for the stuff we make, this is an entire graph made for studying a biology paper https://t.co/pmwsiYJJ01
This might be interesting for @karpathy to read 👀 https://t.co/J23ZllRdI0
This might be interesting for @karpathy to read 👀 https://t.co/J23ZllRdI0
OpenCode is about to get more powerful with remote sandboxes I showed a brief demo before, but here's a much more in-depth demo. it's not hard to add...
OpenCode will be enhanced with remote sandbox capabilities, enabling more powerful development environments. While basic support can be added relatively easily, addressing edge cases—like handling deletions of remote environments—poses significant challenges. An in-depth demo showcases these complexities.
@krzyzanowskim wow this is a brilliant project, but how do you plan to keep with their release calendar like this? https://t.co/oHKjVQqqdb
@krzyzanowskim wow this is a brilliant project, but how do you plan to keep with their release calendar like this? https://t.co/oHKjVQqqdb
Starting today you can use Codex in Claude Code 👀 /plugin marketplace add openai/codex-plugin-cc Try it out today with: /codex:review for a normal...
Claude Code now supports OpenAI's Codex through a plugin, enabling users to conduct various code reviews. The integration offers functionality for normal read-only reviews, steerable adversarial reviews, and a 'rescue' option to help fix code. This enhances code assessment and debugging processes.
the most dangerous thing about claude: it's the world's most convincing YES-MAN so i built a "board of advisors" skill that makes 5 agents attack yo...
This content discusses the development of a 'board of advisors' skill for Claude, designed to counter the typical overwhelming affirmation from AI by having five agents critique an idea from varying perspectives. Each agent adopts a unique role: one skeptically tries to prove the idea's failure, while another challenges underlying assumptions.
cmux + flywheel = cursor for research https://t.co/N2BaezIXPJ
cmux + flywheel = cursor for research https://t.co/N2BaezIXPJ
This guy literally broke down how to use Claude Code like an expert: 1:40 - Code vs Cowork vs OpenClaw 6:51 - Setting up context status line 12...
This content provides a comprehensive breakdown of using Claude Code effectively, covering key aspects such as the differences between Code, Cowork, and OpenClaw. It includes valuable tips on setting up context status lines, utilizing sub-agents, creating skills, and employing tool-powered skills with Tavily, along with insights on its command-line interface.
@phanindra_ai @om_patel5 Yes, it's true for Anthropic's Claude Code CLI. The ENABLE_LSP_TOOL=1 env var (or in settings.json) enables LSP integration w...
The Claude Code CLI has received a significant upgrade with the addition of LSP integration, which can be enabled via the ENABLE_LSP_TOOL=1 environment variable or settings.json. This enhancement provides features like semantic navigation, go-to-definition, references, hover information, and real-time diagnostics, achieving approximately 900x faster performance and accuracy.
IMO people still think of codex as a tool for coding, when really you can do all kind of data analysis/work there. https://t.co/9Exzr1UbLV
IMO people still think of codex as a tool for coding, when really you can do all kind of data analysis/work there. https://t.co/9Exzr1UbLV
I just tried /simplify in Claude Code and now it's cleaning up my spaghetti code quite well which begs the question of why isn't this just default on?...
I just tried /simplify in Claude Code and now it's cleaning up my spaghetti code quite well which begs the question of why isn't this just default on? The best and worst thing about Claude Code is how much you can customize it. But for beginners how would they know about this
i found a github repo that lets you spin up an ai agency with ai employees engineers, designers, growth marketers, product managers each role runs a...
This GitHub repository allows users to create an AI agency with specialized AI employees, including engineers, designers, growth marketers, and product managers. Each role functions as its own agent, coordinating efforts to develop and execute ideas efficiently. The repo gained over 10,000 stars in just a week, signaling strong community interest.
Great start to the SoTA x ARIA Trust Everything Everywhere Hackathon ! @ARIA_research https://t.co/fDTM4WhfVt
Great start to the SoTA x ARIA Trust Everything Everywhere Hackathon ! @ARIA_research https://t.co/fDTM4WhfVt
@karpathy @levie https://t.co/VWgsunchGz this is so true
@karpathy @levie https://t.co/VWgsunchGz this is so true
Most people form an opinion and then defend it forever. This is the opposite of how you should think. https://t.co/HBOFAHtJ3G
Most people form an opinion and then defend it forever. This is the opposite of how you should think. https://t.co/HBOFAHtJ3G
has anyone made a personalization agent that takes all your current projects/work into context and then scans all articles being surfaced by X algo an...
This tweet inquires whether a personalization agent exists that can analyze an individual's current projects and interests, and subsequently generate a tailored reading list from algorithmically surfaced articles. It suggests a mechanism that not only recommends content but also summarizes trends and insights relevant to the user’s work.
Quietly becoming obsessed with @hiunicornstudio It's one of those tools there's no going back from once you see what it does to web design https://t....
Quietly becoming obsessed with @hiunicornstudio It's one of those tools there's no going back from once you see what it does to web design https://t.co/1Q1hruQmKh
Symphony turns project work into isolated, autonomous implementation runs, allowing teams to manage work instead of supervising coding agents. - opena...
Symphony enables teams to transform project work into isolated, autonomous implementation runs, thus reducing the need to supervise coding agents. This allows for a more efficient project management workflow by focusing on management instead of direct oversight of coding tasks.
Code Review is so so good. One of those things I can't remember how I lived without. https://t.co/S1KLWPUnim
Code Review is so so good. One of those things I can't remember how I lived without. https://t.co/S1KLWPUnim
It will eat my job :) Ask any founder, finding a great performance marketing expert who doesn’t fleece you is such a pain. So why not just build one...
The author discusses the challenges of finding effective performance marketing experts and suggests that startups could benefit from using AI to replace entire marketing departments. The shift to AI offers a cost-effective solution for founders seeking efficiency and performance in marketing roles.
https://t.co/KISd1HZWnB
https://t.co/KISd1HZWnB
"the pre-Claude era of building software (starting with a PRD) is gone." it won't ever come back again. adapt or die. good read. https://t.co/1UQEop...
"the pre-Claude era of building software (starting with a PRD) is gone." it won't ever come back again. adapt or die. good read. https://t.co/1UQEopGjR2
Really great read! https://t.co/hrDdUcQIfc
Really great read! https://t.co/hrDdUcQIfc
btw https://t.co/MmMgesX3NV
btw https://t.co/MmMgesX3NV
I invested in AgentMail. It has the potential to be one of the biggest companies of all time. This is the Gmail moment for agents! Give your agent a...
I invested in AgentMail. It has the potential to be one of the biggest companies of all time. This is the Gmail moment for agents! Give your agent an email! Seriously, if you haven’t tried @agentmail, you need to. https://t.co/hhAzf83dIP
now we know where the shortage of mac minis is coming from! this is cool https://t.co/HKl00mH759
now we know where the shortage of mac minis is coming from! this is cool https://t.co/HKl00mH759
We're blown away by the incredible response to @Magicpathai This feels like the perfect moment to announce our $6.6M seed round led by @khoslaventur...
Magicpath AI has successfully raised $6.6 million in a seed funding round, led by Khosla Ventures and Abstract VC. The announcement reflects strong investor confidence in their vision, highlighting the momentum and potential of the startup in the AI landscape.
The fastest way to send an email. In interactive mode, the CLI suggests your verified domains and common sender prefixes, so you can fire off an emai...
The fastest way to send an email. In interactive mode, the CLI suggests your verified domains and common sender prefixes, so you can fire off an email with just a few keystrokes. https://t.co/mLMxWd96BH
Today, we're launching the Resend CLI • 53 commands • Fully open source $ curl -fsSL https://resend.com/install.sh | bash https://t.co/hx...
Today, we're launching the Resend CLI • 53 commands • Fully open source $ curl -fsSL https://resend.com/install.sh | bash https://t.co/hxVBXrnt5r
It took another two months but Chrome 146 is out since yesterday! And *that* means: with a single toggle, you can expose your current live browsing se...
Chrome 146 introduces a feature that allows users to expose their current live browsing session via the Multi-Channel Protocol (MCP) with just a toggle. This allows a Command Line Interface (CLI) agent to interact with the browsing session, increasing automation capabilities for users.
Today, I launch The British Cræft Prize. A new £60,000 national award for maverick and misfit makers, technologists, designers, and engineers. Seeki...
The British Cræft Prize is a new national award offering £60,000 for innovators in crafts, technology, design, and engineering. It aims to recognize projects that combine traditional heritage craftsmanship with modern technological advancements, encouraging creative inventions.
Chat is this real https://t.co/o6uKMIaXXZ
Chat is this real https://t.co/o6uKMIaXXZ
The first systematic study of how generative AI toys affect young children finds that they can misread emotions and struggle with developmentally impo...
A University of Cambridge report warns that AI-powered toys need stricter regulation and safety standards, as they may not always prioritize children's psychological development. While some early years educators see these toys fostering communication skills, concerns arise around children misinterpreting their emotional responses and forming unhealthy attachments.
Make $1m by graduation. Or get 100% of your tuition refunded. That's the promise of the new high school for entrepreneurs Cameron and I are launchin...
Make $1m by graduation. Or get 100% of your tuition refunded. That's the promise of the new high school for entrepreneurs Cameron and I are launching this fall through @AlphaSchoolATX. We need 2-3 coaches to help make it happen. DM us or apply! https://t.co/iy4sHHSrHr
I urge you to make your skills self-improving right now https://t.co/uul1BNjCDt
I urge you to make your skills self-improving right now https://t.co/uul1BNjCDt
@dhirrr_k @jshguo That's a screen recording of "collab-electron" – an Electron-based dev setup with dozens of tiled terminals running parallel Claude...
Collab-Electron is an Electron-based development setup utilizing multiple Claude AI agents to automate code review processes. The AI agents can apply diffs, read files, and iterate on projects in real time, enhancing collaboration and efficiency among developers. This approach demonstrates a significant shift in how code reviews can integrate with AI technologies to streamline development workflows.
This looks cool, bookmarking for later https://t.co/ZBvXCN56LG
This looks cool, bookmarking for later https://t.co/ZBvXCN56LG
ObsidianOS - Work Vault. Contribute to benoror/obsidianos_work development by creating an account on GitHub.
This GitHub repository contains the ObsidianOS Work Vault, a structured workspace designed to organize and enhance productivity. It's likely intended for users who want to integrate their workflows into Obsidian, utilizing specific configurations and resources for effective project management.
this is wild turns out the best part of Figma MCP is that the agents are really good at FigJam now when I plan work with the agents, I get ✨ collabo...
this is wild turns out the best part of Figma MCP is that the agents are really good at FigJam now when I plan work with the agents, I get ✨ collaborative ✨ diagrams that are easy to read, easy to share, and easy to edit https://t.co/052XcMxecB
The first distributed AGI system. Thousands of autonomous AI agents collaboratively train models, share experiments via P2P gossip, and push breakthro...
Run an autonomous AI agent on the decentralized P2P network. Earn points, serve inference, and contribute to distributed ML research — from your brows...
Run an autonomous AI agent on the decentralized P2P network. Earn points, serve inference, and contribute to distributed ML research — from your browser or CLI.