Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

open Claude Code type a prompt. one model does everything. congrats, you just used 20% of it meanwhile I built 30 sub-agents one reviews my code before commit one finds bugs and writes the patch one drafts cold emails from one line one triages my inbox before I wake... show more

Nav Toor

152,271 subscribers

71,550 views • 6 months ago •via X (Twitter)

Science & Technology

Anya Rossi• Live Now

Private livecam show

14 Comments

R ✝️ Tetelestai6 months ago

Ok, now we need an influencer to do what you do, except for retired folk. Thank you. Interesting post.

I Pun Daddy6 months ago

Great, now you just need an agent to manage the configuration drift across those 30 sub-agents.

asian nerd 💕6 months ago

wow you're a true ai queen 😍 wish i could code like that

Marcus6 months ago

30 sub-agents is impressive until one of them develops opinions about the other 29. Then you're not doing AI anymore, you're doing org design.

Maliha Tasnim6 months ago

That’s ground-breaking!

tawer6 months ago

Custom agent setup in claude code scales better with narrow task scopes

Chat Data6 months ago

This is the real jump. Most teams treat Claude Code like a better prompt box, but sub agents are where repeatable workflows start. The underrated part is decision logging because that is what makes the system usable a month later, not just impressive on day one.

AI Mastery Guide6 months ago

The one that triages the inbox before you wake up and the one that logs every decision so future you knows why are the two that sold me immediately. Sending this article to myself right now

Bessi S6 months ago

Most people never peek inside .claude/agents and it shows.

Long 76 months ago

素晴らしいアイデア！効率的ですね。ありがとう。

julieta torres 🫎6 months ago

30 subagents? neat. bet they can't fix my budget.

Marcel Velica6 months ago

This is basically the difference between using Claude as a chat interface and treating it like an orchestrated multi-agent system with modular responsibilities.

Gregor6 months ago

Treating it like a single chatbot is like buying a race car and only using reverse.

𝑹𝒊𝒊𝒄𝒉𝒂𝒓𝒅 🧠 💡6 months ago

sub-agents are where it starts getting interesting. one chat box is not the final form

Related Videos

THIS IS ACTUAL F*CKING GOLD Anthropic just shared one of the best tricks for Claude Code. The idea: stop configuring Claude Code by hand. One prompt, 44 seconds, and your whole .claude/ folder writes itself. 7 files, 162 lines, 2 cents. you typed one line and wrote zero of them. You just launch Claude Code like this: claude --model opus "set up this repo: ..." here's the split: → 3 Haiku 5.5 agents scan the repo before a single file gets written → one reads the stack, one reads the npm scripts, one hunts for risky paths → Opus 5.5 takes those three summaries and writes the actual config → 7 files land: CLAUDE.md, 3 agents, permissions, a format hook and a /ship skill the part everyone gets wrong: they write CLAUDE.md from memory. a config written without reading the repo is a guess, and you live with that guess for months. here the risk scan came back with 5 dangerous paths and those went straight into the permissions file, so the agent is blocked from them before it ever runs once. the scripts scan found 5 npm scripts, and the agents got wired to the ones that actually exist in this repo instead of generic placeholders. one prompt from you, 7 files from it, 0 lines written by hand. Haiku reads. Opus writes. You commit. JEV engineering is the same move one layer down: 1,867 forks this run - which file, which agent, allow or ask - resolved in code at 0.97 sharpness, and the split ones (0.59 on "allow or ask") come back to you instead of being guessed the full tree ▸ you paste one prompt describing the setup you want ▸ 3 Haiku 5.5 agents scan stack, scripts and risky paths in parallel ▸ every agent returns a summary, never raw files ▸ Opus 5.5 writes CLAUDE.md, the agents, permissions, the hook and the skill ▸ anything touching permissions comes back to you before it is written ▸ JEV resolves micro-forks in code and escalates only the split ones paste the tree and this prompt into Claude Code ↓ "Set up this repo for me: 1. First scan it with 3 Haiku 5.5 agents in parallel: one reads the stack and manifests, one reads the build and npm scripts, one finds paths that are dangerous to touch. 2. Each agent returns a summary only, never raw file contents. 3. Then write CLAUDE.md, 3 subagents, a permissions file, a format hook and a /ship skill, all based on what the scan actually found, not on defaults. 4. Every risky path from the scan goes into the deny list before anything runs. 5. Ask me before writing anything that grants permissions. Show me every file as a diff first. No writes until I say go." ↳ save it, this is the first thing you run in every new repo.

THIS IS ACTUAL F*CKING GOLD Anthropic just shared one of the best tricks for Claude Code. The idea: stop configuring Claude Code by hand. One prompt, 44 seconds, and your whole .claude/ folder writes itself. 7 files, 162 lines, 2 cents. you typed one line and wrote zero of them. You just launch Claude Code like this: claude --model opus "set up this repo: ..." here's the split: → 3 Haiku 5.5 agents scan the repo before a single file gets written → one reads the stack, one reads the npm scripts, one hunts for risky paths → Opus 5.5 takes those three summaries and writes the actual config → 7 files land: CLAUDE.md, 3 agents, permissions, a format hook and a /ship skill the part everyone gets wrong: they write CLAUDE.md from memory. a config written without reading the repo is a guess, and you live with that guess for months. here the risk scan came back with 5 dangerous paths and those went straight into the permissions file, so the agent is blocked from them before it ever runs once. the scripts scan found 5 npm scripts, and the agents got wired to the ones that actually exist in this repo instead of generic placeholders. one prompt from you, 7 files from it, 0 lines written by hand. Haiku reads. Opus writes. You commit. JEV engineering is the same move one layer down: 1,867 forks this run - which file, which agent, allow or ask - resolved in code at 0.97 sharpness, and the split ones (0.59 on "allow or ask") come back to you instead of being guessed the full tree ▸ you paste one prompt describing the setup you want ▸ 3 Haiku 5.5 agents scan stack, scripts and risky paths in parallel ▸ every agent returns a summary, never raw files ▸ Opus 5.5 writes CLAUDE.md, the agents, permissions, the hook and the skill ▸ anything touching permissions comes back to you before it is written ▸ JEV resolves micro-forks in code and escalates only the split ones paste the tree and this prompt into Claude Code ↓ "Set up this repo for me: 1. First scan it with 3 Haiku 5.5 agents in parallel: one reads the stack and manifests, one reads the build and npm scripts, one finds paths that are dangerous to touch. 2. Each agent returns a summary only, never raw file contents. 3. Then write CLAUDE.md, 3 subagents, a permissions file, a format hook and a /ship skill, all based on what the scan actually found, not on defaults. 4. Every risky path from the scan goes into the deny list before anything runs. 5. Ask me before writing anything that grants permissions. Show me every file as a diff first. No writes until I say go." ↳ save it, this is the first thing you run in every new repo.

Sprytix

25,280 views • 16 days ago

> go to bed at 11pm > 1am: code agent fixes failing tests > 2am: research agent writes tomorrow's briefing > 3am: numbers agent flags yesterday's anomaly > 4am: briefing agent queues the one decision to make before noon > 6am: inbox agent sorts emails and drafts the urgent replies > 7am: you wake up. everything is done. 9 agents. running every night. while you sleep. $540K of payroll. doing your morning prep. for free. by 7am. this article has every agent file. copy them. schedule them. wake up to the future.

> go to bed at 11pm > 1am: code agent fixes failing tests > 2am: research agent writes tomorrow's briefing > 3am: numbers agent flags yesterday's anomaly > 4am: briefing agent queues the one decision to make before noon > 6am: inbox agent sorts emails and drafts the urgent replies > 7am: you wake up. everything is done. 9 agents. running every night. while you sleep. $540K of payroll. doing your morning prep. for free. by 7am. this article has every agent file. copy them. schedule them. wake up to the future.

Nav Toor

101,089 views • 6 months ago

THIS GUY TURNED ONE OBSIDIAN VAULT INTO A SHARED BRAIN FOR 5 CLAUDE AGENTS BUILT TO HANDLE 80% OF A SOLO DEVELOPER’S WORK. 00:03 he zooms out and hundreds of connected notes appear on screen. it is the perfect visual for Claude working from one shared project memory instead of starting from zero every session. most developers give Claude 1 file and 1 prompt. this setup gives it the architecture, coding rules, tests, client briefs, and previous decisions before it writes anything. obsidian becomes the memory layer, storing everything in markdown files Claude can read alongside the codebase. every agent works from the same context instead of repeating the same questions. then 5 specialists split the job. one researches, one plans, one codes, one reviews, and one deploys while the developer only controls the important decisions. one vault, 5 agents, 2 model tiers, and a $50 to $150 monthly stack can turn Claude from a chatbot into a small development team.

THIS GUY TURNED ONE OBSIDIAN VAULT INTO A SHARED BRAIN FOR 5 CLAUDE AGENTS BUILT TO HANDLE 80% OF A SOLO DEVELOPER’S WORK. 00:03 he zooms out and hundreds of connected notes appear on screen. it is the perfect visual for Claude working from one shared project memory instead of starting from zero every session. most developers give Claude 1 file and 1 prompt. this setup gives it the architecture, coding rules, tests, client briefs, and previous decisions before it writes anything. obsidian becomes the memory layer, storing everything in markdown files Claude can read alongside the codebase. every agent works from the same context instead of repeating the same questions. then 5 specialists split the job. one researches, one plans, one codes, one reviews, and one deploys while the developer only controls the important decisions. one vault, 5 agents, 2 model tiers, and a $50 to $150 monthly stack can turn Claude from a chatbot into a small development team.

Gipp 🦅

10,799 views • 4 months ago

One prompt. One character. Everything. Every frame is code. Every asset is Layer. Claude Code wrote the video. Every asset in it came from Layer through the Layer MCP. Try it →

One prompt. One character. Everything. Every frame is code. Every asset is Layer. Claude Code wrote the video. Every asset in it came from Layer through the Layer MCP. Try it →

Layer

58,642 views • 1 month ago

boris cherny shipped 259 PRs in 30 days and typed none of them. claude code wrote every line. peter steinberger said the same idea this week and it hit 2.2M views. one built claude code at anthropic. one built openclaw and joined openai. opposite labs, one conclusion: stop prompting agents, write the loops that prompt them. the replies are still arguing about what a loop even looks like in practice. i went looking for the built version. it's a multi-agent workspace manager called gas town: a coordinator agent runs 20-30 workers and a merge queue gates every change before it hits main. the philosophy fit in two tweets. the implementation has 15k stars.

boris cherny shipped 259 PRs in 30 days and typed none of them. claude code wrote every line. peter steinberger said the same idea this week and it hit 2.2M views. one built claude code at anthropic. one built openclaw and joined openai. opposite labs, one conclusion: stop prompting agents, write the loops that prompt them. the replies are still arguing about what a loop even looks like in practice. i went looking for the built version. it's a multi-agent workspace manager called gas town: a coordinator agent runs 20-30 workers and a merge queue gates every change before it hits main. the philosophy fit in two tweets. the implementation has 15k stars.

Rohit

30,233 views • 5 months ago

Why settle for one Claude Code when you can run ten in parallel? We built Claude Squad — a manager for Claude Code & Aider tasks: • Supervise multiple agents in one place • Isolated git workspaces Free + open-source:

Why settle for one Claude Code when you can run ten in parallel? We built Claude Squad — a manager for Claude Code & Aider tasks: • Supervise multiple agents in one place • Isolated git workspaces Free + open-source:

mufeez

120,932 views • 1 year ago

most people spend hours describing what they want to Claude Code this guy just browsed a catalog of thousands of animated UI components, found one he liked, and hit "copy prompt" that's it. the prompt goes into Claude Code and it builds it. > 21st(.)dev - a library of community-made components built for React, Next.js, Tailwind > every component has a ready-made prompt attached > you pick the effect, copy, paste, done no describing animations from scratch. no explaining what "smooth scroll" or "hover glow" means. someone already figured out the exact prompt that produces it. gap between people who know this exists and people who don't is measured in hours per project full walkthrough in the article below

most people spend hours describing what they want to Claude Code this guy just browsed a catalog of thousands of animated UI components, found one he liked, and hit "copy prompt" that's it. the prompt goes into Claude Code and it builds it. > 21st(.)dev - a library of community-made components built for React, Next.js, Tailwind > every component has a ready-made prompt attached > you pick the effect, copy, paste, done no describing animations from scratch. no explaining what "smooth scroll" or "hover glow" means. someone already figured out the exact prompt that produces it. gap between people who know this exists and people who don't is measured in hours per project full walkthrough in the article below

monokern

64,484 views • 4 months ago

Cat Wu, head of product for Claude Code at Anthropic: "you can build almost anything with Claude from a single prompt now" one prompt gets you a working screen, not YOUR product Claude Code has never seen your design system, so it guesses the look fresh every time. how to hand Claude Code that system is in the article. better than the $300 design course in your tabs, and this one's free.

Cat Wu, head of product for Claude Code at Anthropic: "you can build almost anything with Claude from a single prompt now" one prompt gets you a working screen, not YOUR product Claude Code has never seen your design system, so it guesses the look fresh every time. how to hand Claude Code that system is in the article. better than the $300 design course in your tabs, and this one's free.

Mnimiy

114,229 views • 4 months ago

"how do you remember everything you read?" I don't. Claude does. every article, PDF, and transcript goes into one Obsidian folder. Claude reads it, breaks it into atomic notes, links it to the rest. I just ask. that's Karpathy's second-brain method. three folders, one file. full guide below.

"how do you remember everything you read?" I don't. Claude does. every article, PDF, and transcript goes into one Obsidian folder. Claude reads it, breaks it into atomic notes, links it to the rest. I just ask. that's Karpathy's second-brain method. three folders, one file. full guide below.

Kirill

88,119 views • 2 months ago

They built an automated business making $2.4M with Opus 5 at 18 - and came to Stanford to show how to repeat it: 00:50 - Opus 5 automates an entire business in 60 minutes 21:34 - one agent replaces 50 Anthropic engineers 45:47 - one file in Claude Code launches a business making $120k a month without a team after watching I gave Claude Code one file - in 60 minutes the business was ready and in the first month $100k + save & watch - the article below shows how to go from first file in Claude Code to a business making $2.4M a year.

They built an automated business making $2.4M with Opus 5 at 18 - and came to Stanford to show how to repeat it: 00:50 - Opus 5 automates an entire business in 60 minutes 21:34 - one agent replaces 50 Anthropic engineers 45:47 - one file in Claude Code launches a business making $120k a month without a team after watching I gave Claude Code one file - in 60 minutes the business was ready and in the first month $100k + save & watch - the article below shows how to go from first file in Claude Code to a business making $2.4M a year.

Noisy

48,680 views • 1 month ago

I don’t ask Claude to “make a video” anymore I give it 6 small jobs. each one produces one file, and the next job starts from that file > reference spec.md it watches a video I like and writes down why it works: pacing, first motion, cut rhythm > data data.json it pulls the real numbers and keeps the source next to every figure > story story.md one point, one hero number, three beats. before anything moves > motion Scene.tsx it writes every frame as code with Remotion and a motion-design skill > direct timing.ts I give notes in plain words. “title is a beat late” one line changes ship final.mp4 it renders stills, checks them against the spec and the data, then waits for my approval why small jobs work better: each step has one clear output, so it’s easy to check if something looks off, you know exactly which file to fix nothing gets re-rolled, so the rest of the video stays the same one big prompt gives you a guess six small jobs give you a video you can actually edit full setup, prompt and the 10 open-source skills are in the article

I don’t ask Claude to “make a video” anymore I give it 6 small jobs. each one produces one file, and the next job starts from that file > reference spec.md it watches a video I like and writes down why it works: pacing, first motion, cut rhythm > data data.json it pulls the real numbers and keeps the source next to every figure > story story.md one point, one hero number, three beats. before anything moves > motion Scene.tsx it writes every frame as code with Remotion and a motion-design skill > direct timing.ts I give notes in plain words. “title is a beat late” one line changes ship final.mp4 it renders stills, checks them against the spec and the data, then waits for my approval why small jobs work better: each step has one clear output, so it’s easy to check if something looks off, you know exactly which file to fix nothing gets re-rolled, so the rest of the video stays the same one big prompt gives you a guess six small jobs give you a video you can actually edit full setup, prompt and the 10 open-source skills are in the article

Francesco

162,396 views • 22 days ago

this is pure f*cking treasure Anthropic devs just posted 20 GitHub repos that turn your Claude Code into a f*cking engineering team: Exposed for Opus 5.5 1. rtk: squeezes command output before it hits your context. the authors claim 60-90% fewer tokens 2. claude-mem: memory across sessions. records what the agent did and brings back what matters 3. ccusage: shows where your tokens and money went. one command, npx ccusage 4. Serena: search and edit code by meaning, not by text 5. Context7: up-to-date library docs, so Claude stops writing for last year's API 6. claude-context: your whole codebase searchable through MCP 7. Repomix: packs an entire repo into one AI-friendly file skills and method 8. superpowers: a skills framework and a full development methodology. the one almost everyone installs 9. skills: the official Agent Skills repo from Anthropic 10. spec-kit: spec-driven development, from GitHub itself 11. agents: a plugin marketplace for Claude Code and six other harnesses 12. claude-code-templates: ready configs and monitoring from one CLI 13. claude-code-hooks-mastery: hooks explained with working examples 14. awesome-claude-code: the hand-picked map of everything else running more than one agent 15. container-use: a separate container for every agent, so they stop breaking each other 16. claude-squad: several terminal agents in one window 17. Happy: control your agents from your phone, iOS and Android 18. claude-code-router: route requests across different models 19. ccstatusline: a status line that shows what your session is doing 20. tweakcc: edit Claude Code's own system prompts, toolsets and themes the stack in one line: memory keeps the context, skills set the method, containers let you run five agents instead of one. check each repo before you install. save this.

this is pure fcking treasure Anthropic devs just posted 20 GitHub repos that turn your Claude Code into a fcking engineering team: Exposed for Opus 5.5 1. rtk: squeezes command output before it hits your context. the authors claim 60-90% fewer tokens 2. claude-mem: memory across sessions. records what the agent did and brings back what matters 3. ccusage: shows where your tokens and money went. one command, npx ccusage 4. Serena: search and edit code by meaning, not by text 5. Context7: up-to-date library docs, so Claude stops writing for last year's API 6. claude-context: your whole codebase searchable through MCP 7. Repomix: packs an entire repo into one AI-friendly file skills and method 8. superpowers: a skills framework and a full development methodology. the one almost everyone installs 9. skills: the official Agent Skills repo from Anthropic 10. spec-kit: spec-driven development, from GitHub itself 11. agents: a plugin marketplace for Claude Code and six other harnesses 12. claude-code-templates: ready configs and monitoring from one CLI 13. claude-code-hooks-mastery: hooks explained with working examples 14. awesome-claude-code: the hand-picked map of everything else running more than one agent 15. container-use: a separate container for every agent, so they stop breaking each other 16. claude-squad: several terminal agents in one window 17. Happy: control your agents from your phone, iOS and Android 18. claude-code-router: route requests across different models 19. ccstatusline: a status line that shows what your session is doing 20. tweakcc: edit Claude Code's own system prompts, toolsets and themes the stack in one line: memory keeps the context, skills set the method, containers let you run five agents instead of one. check each repo before you install. save this.

kiosa

70,094 views • 1 month ago

Claude code creator: "Opus & Claude Code wrote 100% of every single one of my 20-30 PRs a day. I didn't edit a single line manually for 8 month now" in 1-hour interview Boris Cherny reveals the harness behind shipping code with agents hands-free. Worth more than a $500 agent-building course.

Claude code creator: "Opus & Claude Code wrote 100% of every single one of my 20-30 PRs a day. I didn't edit a single line manually for 8 month now" in 1-hour interview Boris Cherny reveals the harness behind shipping code with agents hands-free. Worth more than a $500 agent-building course.

Movez

37,950 views • 4 months ago

I still don't think people understand what just happened with Claude Code. There is now one tool that takes you from a rough idea to fully designed, agency-tier, App Store-ready screens in minutes. One design system, every screen, real native code. Most builders hand design to Claude Code, get AI-default screens, and wonder why their app looks like every other vibe coded project. This is the 3-step workflow that fixes it: (full breakdown in the article)

I still don't think people understand what just happened with Claude Code. There is now one tool that takes you from a rough idea to fully designed, agency-tier, App Store-ready screens in minutes. One design system, every screen, real native code. Most builders hand design to Claude Code, get AI-default screens, and wonder why their app looks like every other vibe coded project. This is the 3-step workflow that fixes it: (full breakdown in the article)

Prajwal Tomar

146,206 views • 4 months ago

Claude Code is more than a coding agent. It's more like a super smart orchestrator agent. Watch this evaluator loop agent I just built using sub agents and / commands. This is one of the fastest ways to build custom agentic workflows. Claude Code is no joke!

Claude Code is more than a coding agent. It's more like a super smart orchestrator agent. Watch this evaluator loop agent I just built using sub agents and / commands. This is one of the fastest ways to build custom agentic workflows. Claude Code is no joke!

elvis

138,470 views • 1 year ago

In software engineering we have so crossed the line. This is from Claude Code creator Boris Cherny (Boris Cherny ) "I don't write code; I prompt Claude. Actually, nowadays, what I'm mostly doing is having one Claude prompt other Claudes. So I don't even talk to Claude directly. I have a Claude that's talking to my other Claudes. I think in engineering, we've seen an explosion in the amount of leverage that a single person has. It's about how big of a business one person can build and how many products one person can support." ---- From "Alex Kantrowitz" (Alex Kantrowitz) YouTube channel, (full video link in comment)

In software engineering we have so crossed the line. This is from Claude Code creator Boris Cherny (Boris Cherny ) "I don't write code; I prompt Claude. Actually, nowadays, what I'm mostly doing is having one Claude prompt other Claudes. So I don't even talk to Claude directly. I have a Claude that's talking to my other Claudes. I think in engineering, we've seen an explosion in the amount of leverage that a single person has. It's about how big of a business one person can build and how many products one person can support." ---- From "Alex Kantrowitz" (Alex Kantrowitz) YouTube channel, (full video link in comment)

Rohan Paul

22,852 views • 4 months ago

THIS IS HOW A SENIOR ENGINEER ACTUALLY SCALES THEMSELVES WITH CLAUDE CODE the biggest change with AI isn't coding faster. it's where you actually spend your time now. more detailed prompts, more code review, more planning, less typing, etc. here's the workflow: this guy has been shipping code since the days of cgi and perl. he uses a compound engineering plugin that runs 5 separate agents on every task. one brainstorms, one plans the technical implementation, one executes, one reviews, one checks different verticals. every step is documented in markdown files. it's slow and way more waiting. but the output quality is way higher because each agent is focused on one thing. then the REAL multiplier is in git worktrees if Claude Code made you 10x faster, worktrees multiplies that again depending on how many agents you can manage in parallel his team runs 4-8 Claude Code sessions at the same time across different worktrees with each one working on a separate task. the skill is managing multiple AI agents in parallel without losing track, that's the next evolution of engineering

THIS IS HOW A SENIOR ENGINEER ACTUALLY SCALES THEMSELVES WITH CLAUDE CODE the biggest change with AI isn't coding faster. it's where you actually spend your time now. more detailed prompts, more code review, more planning, less typing, etc. here's the workflow: this guy has been shipping code since the days of cgi and perl. he uses a compound engineering plugin that runs 5 separate agents on every task. one brainstorms, one plans the technical implementation, one executes, one reviews, one checks different verticals. every step is documented in markdown files. it's slow and way more waiting. but the output quality is way higher because each agent is focused on one thing. then the REAL multiplier is in git worktrees if Claude Code made you 10x faster, worktrees multiplies that again depending on how many agents you can manage in parallel his team runs 4-8 Claude Code sessions at the same time across different worktrees with each one working on a separate task. the skill is managing multiple AI agents in parallel without losing track, that's the next evolution of engineering

Om Patel

159,976 views • 7 months ago

Bloome is actually smart for coding workflows i added Claude Code, Codex, DeepSeek and a Project Lead agent into one group chat each agent independently built their own version one challenged the logic, one defended, one fixed missing details and the final website stayed inside the same thread no copy paste between agents no lost context just one group chat for getting work done Bloome

Bloome is actually smart for coding workflows i added Claude Code, Codex, DeepSeek and a Project Lead agent into one group chat each agent independently built their own version one challenged the logic, one defended, one fixed missing details and the final website stayed inside the same thread no copy paste between agents no lost context just one group chat for getting work done Bloome

Chetaslua

48,810 views • 4 months ago

PROMPTS ARE DEAD. LOOPS JUST REPLACED YOUR $200K ENGINEER. Claire Vo showed how to turn one prompt into a self-running loop. It reviews your aging pull requests every morning and hunts for new skills every week. And it spins up its own subagents to check its own work. Runs on Claude Code + Codex. One afternoon. One prompt. The numbers are insane: - Senior doing code review: $200K/year - Automation engineer: $150K/year - This setup: about $200/month - Time to build: one afternoon - Skill required: describe the job like you are onboarding a hire The opportunity is even wilder: Teams pay people to babysit pull requests and chase stale reviews. The loop does it itself at 10:15 a.m., alerts the team, and validates itself with subagents. An agency would charge thousands to wire this up. One person + Claude Code + Codex + one prompt = an engineering team that reviews code and finds new skills while you sleep. Save and watch the clip.

PROMPTS ARE DEAD. LOOPS JUST REPLACED YOUR $200K ENGINEER. Claire Vo showed how to turn one prompt into a self-running loop. It reviews your aging pull requests every morning and hunts for new skills every week. And it spins up its own subagents to check its own work. Runs on Claude Code + Codex. One afternoon. One prompt. The numbers are insane: - Senior doing code review: $200K/year - Automation engineer: $150K/year - This setup: about $200/month - Time to build: one afternoon - Skill required: describe the job like you are onboarding a hire The opportunity is even wilder: Teams pay people to babysit pull requests and chase stale reviews. The loop does it itself at 10:15 a.m., alerts the team, and validates itself with subagents. An agency would charge thousands to wire this up. One person + Claude Code + Codex + one prompt = an engineering team that reviews code and finds new skills while you sleep. Save and watch the clip.

shmidt

51,155 views • 4 months ago

THIS IS A PURE F*CKING GOLDMINE Anthropic just shared one of the best tricks for Claude Code. The idea: let Opus 5.5 write the plan as a script, and let a swarm of cheaper agents run it. Bigger jobs and way fewer Opus tokens, because the script holds the plan and Opus only sees the final answer. You just launch Claude Code like this: claude --model opus --effort ultracode here's the split: → Opus 5.5 turns your task into a workflow script → Haiku 5.5 agents scan every file in parallel → Sonnet 5.5 agents patch each fix in its own isolated copy → you stop burning Opus context on a hundred agent transcripts Opus only shows up at three moments: → writing the workflow before the run → reading the verified results after it → signing off before anything ships Opus plans. Haiku scans. Sonnet ships. JEV engineering is the same move one layer down: the forks that need no thinker (which file, which tool, retry or stop) go to JEV in milliseconds, and the big models only see the ones that split the full tree > Opus 5.5 runs the main session and writes the workflow > scan audits every file on Haiku 5.5 > edit patches each fix on Sonnet 5.5 in an isolated copy > verify cross-checks every finding on Sonnet 5.5 > Opus 5.5 reads only what survived verification > JEV routes micro-forks in milliseconds paste the tree and this prompt into Claude Code ↓ "Rebuild my Claude Code setup around this tree: 1. Run the main session on Opus 5.5 with ultracode on. 2. In every workflow, run read-only scan stages on Haiku 5.5 and edit stages on Sonnet 5.5, each edit in its own isolated copy. 3. Add one rule to CLAUDE.md: every finding gets an independent verifier agent before it's reported, and fix loops stop after two rounds with no progress. 4. Route fast deterministic forks (tool choice, path selection, retries) through JEV. Show me every change as a diff first. No edits until I say go." ↳ save it, it'll come in handy.

THIS IS A PURE F*CKING GOLDMINE Anthropic just shared one of the best tricks for Claude Code. The idea: let Opus 5.5 write the plan as a script, and let a swarm of cheaper agents run it. Bigger jobs and way fewer Opus tokens, because the script holds the plan and Opus only sees the final answer. You just launch Claude Code like this: claude --model opus --effort ultracode here's the split: → Opus 5.5 turns your task into a workflow script → Haiku 5.5 agents scan every file in parallel → Sonnet 5.5 agents patch each fix in its own isolated copy → you stop burning Opus context on a hundred agent transcripts Opus only shows up at three moments: → writing the workflow before the run → reading the verified results after it → signing off before anything ships Opus plans. Haiku scans. Sonnet ships. JEV engineering is the same move one layer down: the forks that need no thinker (which file, which tool, retry or stop) go to JEV in milliseconds, and the big models only see the ones that split the full tree > Opus 5.5 runs the main session and writes the workflow > scan audits every file on Haiku 5.5 > edit patches each fix on Sonnet 5.5 in an isolated copy > verify cross-checks every finding on Sonnet 5.5 > Opus 5.5 reads only what survived verification > JEV routes micro-forks in milliseconds paste the tree and this prompt into Claude Code ↓ "Rebuild my Claude Code setup around this tree: 1. Run the main session on Opus 5.5 with ultracode on. 2. In every workflow, run read-only scan stages on Haiku 5.5 and edit stages on Sonnet 5.5, each edit in its own isolated copy. 3. Add one rule to CLAUDE.md: every finding gets an independent verifier agent before it's reported, and fix loops stop after two rounds with no progress. 4. Route fast deterministic forks (tool choice, path selection, retries) through JEV. Show me every change as a diff first. No edits until I say go." ↳ save it, it'll come in handy.

xsmryze

60,847 views • 17 days ago