Most agent architectures add a reviewer subagent to catch mistakes. You pay the cost twice: once to do the work, again to check it. Claude Skills flip this. They load domain expertise exactly when the agent needs it—during execution, not after. Real-time guidance, zero review ov…↗

Our Applied AI team with a great write up on improving Claude’s frontend design via Skills. Also with a Claude Code plugin that packages up the skill! claude.com/blog/improving…↗

Pro tip: you can just whitelabel OpenRouter!↗

> its here seconds0.substack.com/p/heres-whats-…↗

BREAKING: Inside Palantir with Alex Karp Exclusive interview in Palantir’s office. We cover: - $PLTR record earnings - ~$500B market cap milestone - New Biography: The Philosopher in the Valley - The rise of AIP & its impact on global enterprise - Morality, value creation, & he…↗

@bentossell @KirkMarple 🥳 also made a thing with droid that circumvents the need for agent CLIs to have a first-party Skill tool: github.com/ain3sh/openski…↗

wow working with @FactoryAI is amazing finished a lot of small features for alprina.com↗

Gemini CLI extensions have led me to contribute to some pretty cool orgs on GitHub over the last few weeks. Which MCP servers do you use the most often? I'll turn them into a Gemini CLI extension so they have enhanced context and custom commands! P.S. Open-source rules! 👍↗

This dynamic will persist until inference moves on device and marginal cost of a prompt is zero. Only then will the AI app ecosystem bloom↗

Michael Truell: How Cursor Builds at the Speed of AI In only a few years, Cursor's AI dev tools have made a name for themselves as fast-shipping and fast-growing. CEO Michael Truell sat down with a16z's Martin Casado to discuss how Michael got here and what it's like to operate…↗

💥NEW AI CODING TUTORIAL: 11 cracked Claude Code 2.0 features to add to your AI coding workflow today TIMESTAMPS 0:00 @claudeai 2.0 is amazing, but... 2:17 Section 1: Must Know Features 2:23 New Models 5:05 Explore Subagent 6:24 Checkpoints and Rewind 7:54 Refreshed Terminal Int…↗

Free, open source. Built it for myself, works for anyone who needs it. Might have some bugs, dunno yet. 🔗 github.com/blacklogos/thr…↗

*gets up on soap box* With the announcement of this new "code mode" from Anthropic and Cloudflare, I've gotta rant about LLMs, MCP, and tool-calling for a second Let's all remember where this started LLMs were bad at writing JSON So OpenAI asked us to write good JSON schemas…↗

@thedanifrim @FactoryAI Yeah, I like it too, Factory Droid’s one of my go-to tools. It’s just a bit more consistent than Claude Code, and they’re clearly refining the agent side fast. I even made a video on it: youtu.be/nc0zg_wAKX0↗

Check out employee of the month over here… @FactoryAI @Splashtop helping me keep tabs on ‘em↗

How to use Cursor (with no coding experience) x.com/i/broadcasts/1…↗

New project live: changelogs.live Still have to set a few things, but I think it is good enough to go live.↗

Cursor put out a blog today stating that semantic search beats grep Semantic search doesn't have to be complicated, and thats exactly why I built SemTools -- to provide agents with a "fuzzy semantic grep search" Semtools github.com/run-llama/semt… Blog cursor.com/blog/semsearch↗

Reading arXiv papers to keep up with AI, but struggled with knowledge recall. Built a mobile app to manage my paper library—adding notes, NotebookLM-style FAQs, and flashcards for better retention. Kudos @FactoryAI !! The experience has been great!↗

What's your favorite Claude Skill?↗

Todos in Gemini CLI are coming together nicely👀 ✅ Completed task ⏩ In progress task ☐ Not started task Todos are still experimental, give them a try by adding "useWriteTodos": true in your settings.json file and let us know what you think! I have found they help keep the mo…↗

TIL .docx, .xlsx, and .pptx are just .zip archives with mostly xml inside.↗

and now you can use bun x to see how this would work for you.↗

The End of Cloud Inference docs.google.com/document/d/1Hl…↗

Today, a part of what I’ve been building over the last few days goes live for beta testing — Chessiro v0.1 ♟️ Chessiro helps you learn chess using your own games — with personalized insights powered by your online chess data. 👇↗

The easiest way to prototype a general agent is to use Claude Code. Write some scripts to access your APIs, write a custom Claude MD, then prompt Claude Code with some queries. Does it work? Use the Claude Agent SDK to productionize it.↗

with cursor 2.0, you can now share commands and rules with links! here are 5 commands from the cursor team:↗

Why even non-coding agents need bash I've done dozens of calls with companies making general agents over the past few weeks and my advice generally boils down to: "use the bash tool more" Here's a concrete example from my email agent:↗

I finally got around to making a tool I've wanted for a long time: you can basically think of it as being "like Gmail for coding agents." If you've ever tried to use a bunch of instances of Claude Code or Codex at once across the same project, you've probably noticed how annoyin…↗

steal my claude commands: github.com/jxnl/dots/tree… also works in @cursor_ai↗