jjs

@theo “Workflow finished - 61 agents” seems oddly specific, that’s all I’m saying

@theo one prompt doesnt really mean anything does it? I can tell claude to spin up 4,000 subagents in one sentence and it would probably do it?

Resubbed at the $100 tier to try 4.8 and the new "ultracode" feature. I hit my limits in a single prompt.

@maytheraven @shubhamJReacts @LowLevelTweets bun install doesn’t run dependency lifecycle scripts (preinstall, postinstall, etc) by default which is exactly what attacks like this exploit

@shubhamJReacts @LowLevelTweets How is it different?

nah im just not gonna run npm install anymore

@JasonL_Capital This company only has 4 employees looks like an extremely risky bet more than an investment

4. $CRML - Critical Metals Corp Controls 92.5% of the Tanbreez project in Greenland - one of the largest heavy rare earth deposits outside of China. Greenland approved the majority ownership transfer in April 2026 and the stock jumped 35% on the news. Initial production targeting 85,000 tonnes of rare earth oxides per year, scalable to 425,000 tonnes. Texas Capital initiated coverage with a Buy rating and a $20 price target. This is the Western world's answer to Chinese rare earth dominance at scale.

Every humanoid robot needs up to 40 motors, and nearly all of them run on rare earth permanent magnets. No magnets, no motors. No motors, no robots. Here are the top 5 stocks you need to know about that address this bottleneck: 1. $MP - MP Materials The only fully integrated rare earth mine-to-magnet operation in the US. Mountain Pass mine in California produces roughly 15% of global rare earth concentrate. Their Fort Worth Independence facility is already producing up to 1,000 metric tons of NdFeB magnets per year, with a $1.25B expansion campus (10X) in Northlake, Texas targeting 10,000 MT total capacity by 2028. GM is a confirmed customer.

@Kekius_Sage It’s the speed it takes to compute and render the next frame of our existence. Machine running our simulation can’t go any faster

Why does light travel at exactly that speed?

@jfikrat @thdxr Agents are intelligence bombs you can throw at anything. executing, architecture etc. Nothing stopping you from using intelligence to your advantage during the design phase

@thdxr Agents are executors, not architects. The start of a project is constraint navigation — you're holding ambiguity on purpose, sketching interfaces that might not survive the next hour. Agents want to resolve everything immediately. That's the opposite of what early design needs.

working on a new project and once again i do not understand how people use agents at the start of the project i didn't even think to, just naturally went around defining interfaces to think through what i'm doing agent comes later

@jeffzwang triage subagents to scan the surface (I have a skill called /hyperscan for this). Create a multi file plan (planning-with-files skill). Iterate on the plan and then let cc implement while tracking progress on the plan files. This usually works. Last resort I ralph loop it

does anyone have any tips on how to prompt/plan when trying to oneshot large projects, like 50K+ LOC?

@lowvisiondave vercel is down bro. new deployments stuck in initializing?

A new toy for y'all to play with this weekend

@PrajwalTomar_ All of this is quite unnecessary. literally just wite the openai codex subscription via oauth. your spend is capped at $200/mo

Most people try OpenClaw and quit after seeing the bill. I almost did too. 3 days in, I was burning $35-40/day. Then I discovered the settings that 90% of users miss. Here's how to run OpenClaw without going broke ↓