Mr. Metrix

@danshipper Had the same issue but solved it with a custom rule: each new thread first reads a source-of-truth md file and the implementation-changelog md file, which must be updated after every implementation change depending on different criteria.

this is what my codex chat history looks like. we absolutely need a new UX for coding agent guis a few thoughts: 1. my usual workflow is i have one main chat where im doing all my work for the day and then separate windows for each individual issue / feature. once those are in good shape and reviewed, i send them to my main chat to qa on staging and deploy 2. the "main chat" paradigm is nice because i want one place to have a good idea of "here's everything i need to accomplish today, what's been done, and what hasn't"—but right now it's at the same level of prominence in the UX / priority as every other chat which isn't great. 3. the main chat "orchestrator" should also be able to easily know what's going on in other threads. i need to be able to reference other chats easily and also have each sub thread have some idea of what the others are doing by default (should be able to isolate if i want though) 4. it's pretty wild to be doing your n-th codex run of the day, type what you want, and then have the model be like "okay, let's see what's going on here, ahhh there's a repo, and it has X features..." as if it has no idea what i've been doing that day. seems like a huge waste of time and tokens, we need something better by default.

@shadcn Legend 💯

Today, we're releasing shadcn/cli v4. It packs a ton of features: shadcn/skills, presets, dry-run, monorepo and more. If you're using shadcn/ui with coding agents or need better control over the defaults, this is for you. Here's everything new:

here's my cursor usage during 2025! go see yours at: cursor.com/2025

I've been using GPT-5.1 Codex Max High Fast in @cursor_ai for the past few days (both for Planning and Implementation) and it's been the best model I've used so far. Very few errors, great at following specific instructions, and much less AI slop being generated in code.

@TimoBuilds_ @creem_io Congrats man

OMG, HAVE TO CRY MY FIRST SALE EVER 🎉 AND IT'S HUUUGE

@ericzakariasson @imrat I’ve been enjoying plan mode. I found it to work better when used with Sonnet 4.5 and then gpt-5-high for implementation, especially if its a more complex task. What would be cool if Plan mode automatically added version history to the plan after iterations or inputs

@imrat i use gpt-5 or sonnet 4.5 in plan mode to plan larger features. then i skim through the plan, iterate on it by hand or with agent, and eventually click "Build"

have your agent run for longer with plan mode in cursor! use shift+tab to toggle between modes for quick access

@Praha37v Honestly, this might just be the best onboarding experience I’ve ever come across. Incredible work 👏

Client asked for a unique onboarding experience, so we did this.

@adamstern_ @diabrowser Ahhh finally 💪

Vertical tab option is available today in @diabrowser! I've been loving switching between top/side tabs depending on the task. Here are a few technical details I'm most excited about…

Came across the same issue while running 5 background agents today. I accidentally started one from the main branch instead of choosing the actual branch I wanted and I spent 10 minutes trying to find it. For some reason it automatically archived it 🤷‍♂️ but that right click was a saving grace

Been using Cursor Background agents intensively for the last week. Here's some of the things I found out that I could not find in the docs.

Fascinating reading, just goes to show how important it’s becoming to properly and comprehensively configure and setup clear boundaries in any type of ‘system prompt’ for any product using AI.

@joshm @diabrowser This will be very handy, got used to vertical tabs in Arc

Vertical tabs in @diabrowser make it feel like AI chat and a web browser actually fused together. There’s something natural about having your browser tabs right next to your AI chats in a sidebar. Rolling out in a week or two! Don’t worry, it’s just an option in Settings.

@serafimcloud @cursor_ai This looks amazing, can’t wait to try it out.

Meet Phion - the coolest way to vibecode in @cursor_ai, making it feel like a no-code tool. It's so simple and fun! Available today!

@Ammar110_SM @MarcelFromMimic This looks stunning

@MarcelFromMimic Medical Dashboard ✨

let’s make a thread of the best design work on the planet. I'll start:

@iannuttall Thanks for taking the time to put this together, I’ve been looking to try out this setup for the past few days and this will be immensely helpful. 🙏

If you want to give me props, I'd appreciate the watch time on YT 🙏 youtube.com/watch?v=U_vwfQ…

How to use Claude Code for beginners (This too way too long to edit!) Watch the 15m video covering: - Setting it up in Cursor - How to add a lot of context (hint: not by copy/pasting) - Covering lots of slash commands - How to paste images + lots more 🔥 YT link in replies ↓

@www_dot_andrew @boltdotnew Congrats man 🎉

@DmytroKrasun Your attitude is spot on man, these cats will just help you in the long run. I knew a guy years ago who would share his own premium wp themes on torrents with enough messy code to cause them headaches and those same ppl would end up buying the actual paid themes from him.

To whoever created that and copied ScreenshotOne.com completely: I am almost at $20K MRR, and I now see new copycats every week. I am not afraid of them. I just don't believe it really works. Every few months I see a few of them abandoned. You will likely spend a few years to realize how hard it is to pull off at least $1K MRR in this niche and how many problems it has. Not gatekeeping, honestly sharing. I love ScreenshotOne and will keep working on it. But if I were to start again, I would not create a screenshot API, I would create something in a huge market and would focus on one simple feature that solves one particular pain well. Then make the product so good, that people would cry from happiness when using it. But if you anyway decide to compete, pick one particular feature and make it better than others, you need to differentiate yourself somehow. Otherwise, why would somebody choose your copy of ScreenshotOne than the original?

@thekitze I laughed way too hard at this 😅

liquid glass vs shadcn

Just want to give a shoutout to the team @todoist (no affiliation, just love their product). I've been using their Pro plan for years and this is one of the apps I use several times a day to organize everything. These guys silently ship updates almost every day 🚀💯

@v0 This is great, especially when you need to edit something specific and don’t want it to generate multiple files just to get it looking like you want it to.

Introducing Design Mode on v0: • Quickly tweak your generation — copy, typography, layout, colors, styling, and more • Preview changes and save when ready • No need to spend credits or wait for an LLM • Tailwind and shadcn native

@daniel_dhawan This was highly needed, looking forward to trying out the integration with Cursor, that was the missing piece for me.

You can now use Cursor & Rork via GitHub integration Or hire a Pro engineer if you want help in polishing your app This was one of the most requested features!