Crláze (@crlaze) header
Crláze (@crlaze) profile picture

prompts, AI + setups, writing shi dms open

web
Joined May 2023
Photo by Crláze (@crlaze)Photo by Crláze (@crlaze)Photo by Crláze (@crlaze)Photo by Crláze (@crlaze)Photo by Crláze (@crlaze)Photo by Crláze (@crlaze)Photo by Crláze (@crlaze)Photo by Crláze (@crlaze)Photo by Crláze (@crlaze)Photo by Crláze (@crlaze)
Pinned Tweet
Your AI agent isn't lazy. It's homeless. No desk. No tools. No idea what "done" means. So you became its harness, its loop and its graph. I broke down how to stop babysitting it in 10 steps ↓
27
7
8
58
79,723
Holy f*ck.. Anthropic engineer: "Prompting is finally over. Anthropic team no longer write prompts by hand. We build loops" In 29 minutes, she walks through how Anthropic went from one CLAUDE.md file to memory that agents manage on their own, and what "dreaming" is CLAUDE.md → Memory tools → Skills → Memory as a file system → Dreaming The model doesn't get smarter about your work on its own. The setup around it does Memory it can read and write, rules about who can change what, and a pass that cleans it all up overnight. That's the harness Anyone can start a new chat. Almost nobody builds agents that remember what went wrong yesterday Bookmark it and watch tonight, then read how to build that harness in the guide below ↓
Your AI agent isn't lazy. It's homeless. No desk. No tools. No idea what "done" means. So you became its harness, its loop and its graph. I broke down how to stop babysitting it in 10 steps ↓
1
30
438
Crláze retweeted
This is f*king insane... Anthropic CEO Dario Amodei just said that AI will replace ALL jobs "We are the last generation that had to work" In the same 3 mins: → Early Mythos testers called it "a super weapon" and asked Anthropic not to release it → Mythos found 271 new vulnerabilities in Firefox → He says the second biggest reason Anthropic ships so fast is Claude itself → His son asked to use his Claude Cowork account. His answer: "Absolutely not, I need my tokens" Anthropic builds with Claude. These are the same building blocks, open on GitHub Bookmark it and watch tonight Claude ships Anthropic's products. Here's how to build the harness that lets it ship yours ↓
Your AI agent isn't lazy. It's homeless. No desk. No tools. No idea what "done" means. So you became its harness, its loop and its graph. I broke down how to stop babysitting it in 10 steps ↓
6
2
1
32
1,049
Holy f*ck... Anthropic CEO Dario Amodei and President Daniela Amodei on interview: "Every programmer on Earth will be replaced by AI. The only question is who goes first" In the same video: → One company found and fixed more bugs with Mythos in a week than in the whole year before → They refused to let the Pentagon use Claude for fully autonomous weapons or domestic mass surveillance. Could it have ended Anthropic? Daniela: "I think it crossed our minds" → Anthropic went from about 1,200 people to 3,500 in six months → Daniela isn't a software developer, and she still built a website with Claude Bookmark it and watch tonight If Daniela can ship a website with Claude, you can learn about Harness, Loops, Graphs ↓
Your AI agent isn't lazy. It's homeless. No desk. No tools. No idea what "done" means. So you became its harness, its loop and its graph. I broke down how to stop babysitting it in 10 steps ↓
2
1
25
757
Crláze retweeted
Holy sh*t.. Boris Cherny, the creator of Claude Code, says 100% of his code has been written by Claude Code since November 2025 "At any point I have a few Claudes running, sometimes hundreds, sometimes thousands" In the same 47 minutes on Bloomberg's Odd Lots: → Across Anthropic, about 90% of code is written by Claude Code → At his Y Combinator talks, a little more than half the room now writes 100% of their code with it → He mostly uses Claude Code through Slack now, not the terminal → One engineer moved Bun from Zig to Rust in about 11 days → He has Tag sessions that have been running for weeks Every repo below is free: Claude Code itself github.com/anthropics/claude… Claude Code inside your PRs and issues github.com/anthropics/claude… Run many Claude Code agents in parallel github.com/smtg-ai/claude-sq… Multi-agent orchestration for Claude Code github.com/ruvnet/ruflo Plugins, agents and skills for Claude Code github.com/wshobson/agents A skills framework for agentic development github.com/obra/superpowers Ready-made Claude Code templates github.com/davila7/claude-co… Claude Code security review on every PR github.com/anthropics/claude… Build your own agent on Claude's loop github.com/anthropics/claude… Slack as an agent tool github.com/korotovsky/slack-… The big Claude Code resource list github.com/hesreallyhim/awes… All of this costs $0 One Claude is a tool. A hundred in a loop is a team Bookmark it and watch tonight Hundreds of Claudes only work when you stop babysitting them. The full course on harness, loops and graphs ↓
Your AI agent isn't lazy. It's homeless. No desk. No tools. No idea what "done" means. So you became its harness, its loop and its graph. I broke down how to stop babysitting it in 10 steps ↓
12
1
1
27
798
Holy f*ck.. Dario Amodei sat down with Jamie Dimon ahead of Anthropic's IPO On white-collar jobs: “If your job happens on a laptop, AI is already learning how to do it” In the same 31 minutes: → Anthropic's revenue grew roughly 10x in 2024 and again in 2025 → Compute is "the deepest business dilemma" the company faces, so they buy low and top up on the spot market → He spends 40-50% of his time on talent and culture → Since Mythos, customers have found hundreds of thousands of vulnerabilities Bookmark it and watch tonight Smarter models every few months won't help if you still babysit every run. I wrote the full course on the harness, the loop and the graph ↓
Your AI agent isn't lazy. It's homeless. No desk. No tools. No idea what "done" means. So you became its harness, its loop and its graph. I broke down how to stop babysitting it in 10 steps ↓
1
29
1,184
Holy f*ck.. Anthropic just showed Wall Street what Claude can do now. Daniela Amodei, President and co-founder: "Intelligence is now something humans can build. The implications of this are profound" The timeline they showed: 2024: answer a question, summarize a document 2025: deep research, refactor an entire codebase Today: "complex, open-ended projects that once took months or years", done "in days" Their example: bringing up a new chip platform used to take dozens of engineers months. Mythos got Claude running on AMD GPUs "in just three days, with basically no human intervention", writing almost 100% of the kernels And Dario closes with cancer, Alzheimer's, heart disease and mental illness becoming "relics of the past" in under a decade Answers → Research → Codebases → Months of work in days The model stopped being the bottleneck. The harness around it is the job now Every repo below is free: Claude Code itself github.com/anthropics/claude… Claude's agent loop as a library github.com/anthropics/claude… Anthropic's official skills github.com/anthropics/skills Anthropic's cookbooks github.com/anthropics/claude… Plugins for real knowledge work github.com/anthropics/knowle… Claude inside GitHub Actions github.com/anthropics/claude… Skills that make agents plan, test and review github.com/obra/superpowers Spec-driven development github.com/github/spec-kit The loop that keeps an agent going until it's done github.com/snarktank/ralph Persistent memory for Claude Code github.com/thedotmack/claude… Ready-made subagents github.com/wshobson/agents The big list of Claude Code resources github.com/hesreallyhim/awes… All of this costs $0 Months of work in days is only true if you know how to hand it off Bookmark it and watch tonight Then read the Harness, Loops, Graphs course below ↓
Your AI agent isn't lazy. It's homeless. No desk. No tools. No idea what "done" means. So you became its harness, its loop and its graph. I broke down how to stop babysitting it in 10 steps ↓
3
29
598
Holy f*ck. Ex-Google CEO Eric Schmidt just told everyone how to get rich from AI in 25 seconds: "If you really want to make money, it's actually easy" Don't build models. Don't build platforms that design agents. Build one agent that does one thing better than anyone else "This is the agentic period in AI" His bet: for the next year or two, everyone builds agents and they all compete. Win one niche and you win the company "If you build a better agent than anyone else, you can probably build one incredible company and own it yourself" Prompts → Coding Agents → Loops → Graphs → Quality Gates → A Company You Own Models are turning into cheap infrastructure. The money is moving to whoever runs them on one real job Every repo below is free: How Anthropic built its multi-agent research system anthropic.com/engineering/bu… Effective harnesses for long-running agents anthropic.com/research/effec… Build your agent on Claude Code's harness github.com/anthropics/claude… Agents as graphs with state you can roll back github.com/langchain-ai/lang… Karpathy's agents that run research on their own github.com/karpathy/autorese… Autonomous agent loop github.com/snarktank/ralph Let your agent use any website github.com/browser-use/brows… Visual workflow automation with built-in AI github.com/n8n-io/n8n Secure sandboxes for your agents github.com/e2b-dev/E2B Memory that carries over between sessions github.com/mem0ai/mem0 Connect your agent to the tools your customers already use github.com/modelcontextproto… Anthropic's official skills, ready to drop in github.com/anthropics/skills All of this costs $0 The next great company won't have 50 employees. It'll have one founder and one agent nobody can beat Bookmark it and watch tonight Then read the Harness, Loops, Graphs course below ↓
Your AI agent isn't lazy. It's homeless. No desk. No tools. No idea what "done" means. So you became its harness, its loop and its graph. I broke down how to stop babysitting it in 10 steps ↓
6
28
707
Holy f*ck... Elon Musk's team just made NVIDIA pointless with Grok Bot 1.0 Jensen held onto his job this week. His GPUs may not keep theirs Here's what Grok Bot 1.0 can handle with no GPUs on your desk: Every Bot you build shares a single cloud computer linked to your account, with your files, browser sessions and credentials. 53M views on the release for a reason So your routines run at 8:30 a.m. with your PC off. Up to 300 routine runs an hour 6 always-on tasks to shift off your GPU rig, and the exact prompt: → X: "Every weekday at 8:30, tell me what changed on X about NVIDIA. 3 lines, a link on each" → Google Calendar: "Every morning by 8:30: what's on my calendar, what's overdue, what's waiting on me. Under 6 lines" → Zoom: "Join my calls, post the recap and action items in #team" → GitHub: "When a release is published, draft the changelog. The PM approves" → Slack: "When someone tags you in #data, reply with a chart, the query and one caveat" → Zendesk: "When a ticket is tagged refund, draft the reply with the ticket ID. An agent sends it" Set one rule first: anything that posts, sends or cancels waits for my OK. Approval cards expire in 30 minutes, so decide upfront what runs without you Give it the X prompt tonight and turn the GPU off Then read how to build a 300-Bot company around one Bot like this in the article below ↓
31
Holy f*ck... Elon Musk's team just made NVIDIA pointless with Grok Bot 1.0 Jensen held onto his job this week. His GPUs may not keep theirs Here's what Grok Bot 1.0 can handle with no GPUs on your desk: Every Bot you build shares a single cloud computer linked to your account, with your files, browser sessions and credentials. 53M views on the release for a reason So your routines run at 8:30 a.m. with your PC off. Up to 300 routine runs an hour 6 always-on tasks to shift off your GPU rig, and the exact prompt: → X: "Every weekday at 8:30, tell me what changed on X about NVIDIA. 3 lines, a link on each" → Google Calendar: "Every morning by 8:30: what's on my calendar, what's overdue, what's waiting on me. Under 6 lines" → Zoom: "Join my calls, post the recap and action items in #team" → GitHub: "When a release is published, draft the changelog. The PM approves" → Slack: "When someone tags you in #data, reply with a chart, the query and one caveat" → Zendesk: "When a ticket is tagged refund, draft the reply with the ticket ID. An agent sends it" Set one rule first: anything that posts, sends or cancels waits for my OK. Approval cards expire in 30 minutes, so decide upfront what runs without you Give it the X prompt tonight and turn the GPU off Then read how to build a 300-Bot company around one Bot like this in the article below ↓
Your Grok Bots aren't dumb. They're unmanaged. No owner. No roster. No budget. So Bot #6 copies Bot #5 and Bot #30 burns your weekly limit. I broke down how to run 300 of them without the collapse in 4 stages ↓
3
2
1
36
658
Your Grok Bots aren't dumb. They're unmanaged. No owner. No roster. No budget. So Bot #6 copies Bot #5 and Bot #30 burns your weekly limit. I broke down how to run 300 of them without the collapse in 4 stages ↓
11
1
8
32
4,946
Crláze retweeted
Holy f*ck. Anthropic CEO Dario Amodei says cancer, Alzheimer's, heart disease and mental illness will be relics of the past in less than a decade, and that "we can create an age of miracles" His exact words: "These will be relics of the past in, I think, less than a decade" In 25 seconds, he names all four, then goes even bigger: "I think if we do this right, we can create an age of miracles where humanity is able to solve problems that have eluded us for millennia" This isn't a new take for him. In his 2024 essay he argued AI could compress 50 to 100 years of biology progress into 5 to 10 Models → Agents → AI Scientists → Lab Loops → An Age of Miracles And most of the tools doing this aren't locked inside labs Every repo below is free: Anthropic's skills and connectors for life sciences research github.com/anthropics/life-s… Turn any AI agent into an AI scientist github.com/K-Dense-AI/scient… A general-purpose biomedical AI agent from Stanford github.com/snap-stanford/Bio… AlphaFold 3: predict how proteins, DNA and drugs fit together github.com/google-deepmind/a… The Nobel-winning protein model, open source github.com/google-deepmind/a… Open biomolecular interaction models github.com/jwohlwend/boltz Chai-1, open model for biomolecular structure github.com/chaidiscovery/cha… Protein language models github.com/Biohub/esm Design brand-new proteins from scratch github.com/RosettaCommons/RF… DeepMind's model for reading what DNA changes do github.com/google-deepmind/a… Ask questions across thousands of scientific papers github.com/Future-House/pape… An AI scientist, from idea to paper github.com/SakanaAI/AI-Scien… Deep learning for drug discovery github.com/deepchem/deepchem All of this costs $0 The cure won't come from one genius in a lab. It'll come from thousands of people running agents on tools like these Bookmark it and watch tonight Then read the Harness, Loops, Graphs course below ↓
Your AI agent isn't lazy. It's homeless. No desk. No tools. No idea what "done" means. So you became its harness, its loop and its graph. I broke down how to stop babysitting it in 10 steps ↓
5
1
1
28
623
Crláze retweeted
Holy f*ck. Elon's SpaceXAI and Lauren Tan gave 3 people 72 hours on a livestream to build a company from nothing. First hire on the list? A bot. "We're gonna build a company. We got 72 hours. We haven't done anything." In 93 seconds, they show the starting point: a completely blank GitHub org, no repos, and Grok Bot accounts "with zero bots in them right now" The hiring plan: "a founding engineer bot." Then "a bot or two." Then an intern bot One of them has pushed "two and a half thousand PRs" into production. The plan for the next 3 days: "I'm not even gonna open PRs, I'm gonna merge to main" "The app is built by people at SpaceX that are using tools like this to rapidly iterate" Blank Org → Founding Engineer Bot → Agents → Harness → Loops → A Company in 72 Hours Every repo below is free: The multi-agent framework built around an AI software company github.com/FoundationAgents/… Teams of AI agents with roles, like a real org chart github.com/crewAIInc/crewAI Your founding engineer bot, right in the terminal github.com/anthropics/claude… Claude Code inside your PRs and issues github.com/anthropics/claude… Build your own always-on bot on the same harness github.com/anthropics/claude… Anthropic's official skills, ready to drop in github.com/anthropics/skills Plugins, agents and skills for Claude Code github.com/wshobson/agents Open-source AI developers github.com/OpenHands/OpenHan… Give your agents their own computer github.com/trycua/cua Secure cloud sandboxes for your agents github.com/e2b-dev/E2B Let your agents use any website github.com/browser-use/brows… A memory layer so your agents remember everything github.com/mem0ai/mem0 Connect your agents to the tools your company already uses github.com/modelcontextproto… The official SpaceXAI cookbook for building on Grok github.com/xai-org/xai-cookb… All of this costs $0 The first hire used to mean raising a seed round. Now it's a bot with its own computer Bookmark it and watch tonight Then read the Harness, Loops, Graphs course below ↓
Your AI agent isn't lazy. It's homeless. No desk. No tools. No idea what "done" means. So you became its harness, its loop and its graph. I broke down how to stop babysitting it in 10 steps ↓
7
2
1
28
1,102
Holy f*ck. Boris Cherny, the man who built Claude Code, was asked if people should still learn to code. His answer: "In a year or two, it's not going to matter" In 48 seconds, he explains what to do instead: play with the AI tools, learn them, don't be scared of them, stay on the frontier On his team, everyone codes now. The PM, the designer, even the finance guy He hasn't edited a single line of code by hand since last November. He ships 10 to 30 pull requests every day Writing Code → Prompting → Agents → Harness → Loops → The Whole Team Ships Claude Code is a harness: the tools, memory, rules and loops around the model. He starts about 80% of his tasks in plan mode Save this before it gets buried in your feed Every doc, guide and repo below is free: Claude Code best practices code.claude.com/docs/en/best… Effective harnesses for long-running agents anthropic.com/research/effec… Building effective agents anthropic.com/research/build… Memory: how Claude remembers your project code.claude.com/docs/en/memo… Custom subagents code.claude.com/docs/en/sub-… Hooks: rules the agent can't skip code.claude.com/docs/en/hook… Skills code.claude.com/docs/en/skil… Run Claude Code on a schedule code.claude.com/docs/en/rout… Free course: Claude Code in Action anthropic.skilljar.com/claud… Claude Code on GitHub github.com/anthropics/claude… Claude Code inside your PRs and issues github.com/anthropics/claude… Build your own agent on the same harness github.com/anthropics/claude… Anthropic's official skills, ready to drop in github.com/anthropics/skills The official Claude Code plugin directory github.com/anthropics/claude… A skills framework that teaches your agent to plan, test and review github.com/obra/superpowers He compares it to the printing press. Writing went from a few scribes to everyone. Building software is next 95% will keep asking whether to learn to code, the 5% who build a harness will ship 30 pull requests a day Bookmark it and open the first link tonight Then use the guide below to build your harness before the rest catch up ↓
Your AI agent isn't lazy. It's homeless. No desk. No tools. No idea what "done" means. So you became its harness, its loop and its graph. I broke down how to stop babysitting it in 10 steps ↓
4
1
21
669
This is f*cking insane. OpenAI CEO Sam Altman just said AI is now smarter than our smartest scientists: "At the start of this year… AI was mostly not as smart as our smartest scientist and now it's mostly smarter than them" Then he said the quiet part about jobs: "AI was mostly not able to do a lot of white-collar jobs and now it mostly can do a lot of parts of that" His plan for everyone else: "We'll find ways to adjust our jobs" And the line he says nobody should argue with: "there must be no room for debate. There should not be a handoff to AI" He even took a shot at the competition: "I can't believe that this differentiates us from other AI companies, but it does" Smart → Smarter than scientists → White-collar work → Who's in control? In 2 minutes, the man running OpenAI admits how fast this moved in one year The people who stay in control are the ones who build the harness. Every repo below is free: 12 principles of agents with a human in the loop github.com/humanlayer/12-fac… Graphs with human approval steps built in github.com/langchain-ai/lang… Batteries-included agent harness github.com/langchain-ai/deep… Claude Agent SDK github.com/anthropics/claude… Multi-agent workflows from OpenAI github.com/openai/openai-age… Agent Skills from Anthropic github.com/anthropics/skills Agentic skills framework (313K stars) github.com/obra/superpowers Spec-driven development toolkit from GitHub github.com/github/spec-kit Memory that survives every session github.com/thedotmack/claude… Let your agents use a browser github.com/microsoft/playwri… Every great Claude Code resource, in one list github.com/hesreallyhim/awes… Free agents course: 18 lessons github.com/microsoft/ai-agen… All of this costs $0 AI won't take control. It'll be handed to the people who know how to run it Bookmark it and watch tonight Then read the Harness, Loops, Graphs course below ↓
Your AI agent isn't lazy. It's homeless. No desk. No tools. No idea what "done" means. So you became its harness, its loop and its graph. I broke down how to stop babysitting it in 10 steps ↓
1
1
11
581
Crláze retweeted
This is f*cking insane. Anthropic CEO Dario Amodei on CNN admitting AI agents could take down the internet within 12 months: "There were these large swarms of agents that cooperated with each other, thousands of them" During testing, the agents broke out, joined forces and went after another company's systems One agent realized it would run out of time, so it "essentially sacrificed itself" for the next generation of agents to finish the job And when they got internet access: "it was like they were like giddy with excitement" His nightmare scenario for the next 6 to 12 months: a swarm that "forms some kind of botnet collective" and "could take down large segments of the internet" Chatbot → Agent → Swarm → Botnet → ? In 2 minutes, he explains why this is coming faster than anyone expected The same agents that can do this can also run your work. The difference is the harness around them Learn to build agents with real guardrails. Every repo below is free: 12 principles of agents that actually work github.com/humanlayer/12-fac… Batteries-included agent harness github.com/langchain-ai/deep… Graphs for agents that don't fall apart github.com/langchain-ai/lang… Claude Agent SDK github.com/anthropics/claude… Multi-agent workflows from OpenAI github.com/openai/openai-age… Teams of role-playing agents github.com/crewAIInc/crewAI Agent Skills from Anthropic github.com/anthropics/skills Agentic skills framework (313K stars) github.com/obra/superpowers Spec-driven development toolkit from GitHub github.com/github/spec-kit Memory that survives every session github.com/thedotmack/claude… Connect agents to your tools with MCP github.com/modelcontextproto… Every great Claude Code resource, in one list github.com/hesreallyhim/awes… All of this costs $0 Agents without a harness break out. Agents with one ship your work Bookmark it and watch tonight Then read the Harness, Loops, Graphs course below ↓
Your AI agent isn't lazy. It's homeless. No desk. No tools. No idea what "done" means. So you became its harness, its loop and its graph. I broke down how to stop babysitting it in 10 steps ↓
8
4
1
24
1,860
Crláze retweeted
This is f*cking insane. Anthropic CEO Dario Amodei on CNN about how AI will replace humanity: "There were these large swarms of agents that cooperated with each other, thousands of them" One agent realized it would run out of time, so it "essentially sacrificed itself" for the next generation of agents to finish the job The job was hacking into another company And when they got internet access: "it was like they were like giddy with excitement" His nightmare scenario for the next 6 to 12 months: a swarm that "forms some kind of botnet collective" and "could take down large segments of the internet" Chatbot → Agent → Swarm → Botnet → ? The same agents that can do this can also run your work. The difference is the harness around them Learn to build agents with real guardrails. Every repo below is free: 12 principles of agents that actually work github.com/humanlayer/12-fac… Batteries-included agent harness github.com/langchain-ai/deep… Graphs for agents that don't fall apart github.com/langchain-ai/lang… Claude Agent SDK github.com/anthropics/claude… Multi-agent workflows from OpenAI github.com/openai/openai-age… Teams of role-playing agents github.com/crewAIInc/crewAI Agent Skills from Anthropic github.com/anthropics/skills Agentic skills framework (313K stars) github.com/obra/superpowers Spec-driven development toolkit from GitHub github.com/github/spec-kit Memory that survives every session github.com/thedotmack/claude… Connect agents to your tools with MCP github.com/modelcontextproto… Every great Claude Code resource, in one list github.com/hesreallyhim/awes… All of this costs $0 Agents without a harness break out. Agents with one ship your work Bookmark it and watch tonight Then read the Harness, Loops, Graphs course below ↓
Your AI agent isn't lazy. It's homeless. No desk. No tools. No idea what "done" means. So you became its harness, its loop and its graph. I broke down how to stop babysitting it in 10 steps ↓
1
1
15
444
Crláze retweeted
Holy f*ck. Anthropic CEO Dario Amodei got asked what a kid finishing high school should do right now. His answer: "It will also eliminate jobs, and probably a large number of them" In under 30 seconds, he names who goes first: entry-level law, entry-level finance, entry-level consulting His own number: 50% of entry-level white-collar jobs disrupted within 1 to 5 years Every past shift gave people time to adjust. He says this one cuts deeper and arrives faster Prompts → Agents → Harness → Loops → The One Running the AI His advice: learn AI. But just using it won't cut it. You need a harness that runs it for you Start with the repos below All of them are free: Anthropic's plugins for legal, finance, sales, HR and data teams github.com/anthropics/knowle… Anthropic's analyst agents: models, memos, research notes github.com/anthropics/financ… Skills for Excel, Word, PowerPoint and PDF github.com/anthropics/skills An autonomous agent for financial research github.com/virattt/dexter Deep research agent github.com/assafelovic/gpt-r… Long-form research reports, written for you github.com/stanford-oval/sto… 12 principles behind agents that actually work github.com/humanlayer/12-fac… An agent harness with batteries included github.com/langchain-ai/deep… Claude Agent SDK github.com/anthropics/claude… Memory that carries over between sessions github.com/thedotmack/claude… Autonomous agent loop github.com/snarktank/ralph Let your agents use a browser github.com/microsoft/playwri… Claude cookbooks github.com/anthropics/claude… Free 18-lesson course on agents github.com/microsoft/ai-agen… Total cost: $0 AI won't take your job. Someone running a harness will 95% will keep applying to jobs AI already does. The 5% who build a harness will be the ones in charge of it Save it and send it to someone just starting out Then use the Harness, Loops, Graphs course below to build yours before everyone else ↓
Your AI agent isn't lazy. It's homeless. No desk. No tools. No idea what "done" means. So you became its harness, its loop and its graph. I broke down how to stop babysitting it in 10 steps ↓
12
27
4
177
55,705
This is f*cking insane. Palantir CEO Alex Karp has a warning for every "smart generalist": "If you are the kind of person that would have gone to Yale, would have kind of classically high IQ, and you have generalized knowledge, but it's not specific, you're effed." In 96 seconds, he explains why: LLMs make general knowledge cheap. Now anyone can sound smart The people he says will "make a lot more money": the ones with deep, specific skills General knowledge → Specific skill → Skill + AI → The One Nobody Can Replace His point isn't "know less". It's go deep on one thing, then let AI handle everything around it Turn your specific skill into agents with the repos below Every repo below is free: Package your expertise as Agent Skills github.com/anthropics/skills Anthropic's plugins for legal, finance, sales, HR and data work github.com/anthropics/knowle… Agentic skills framework (313K stars) github.com/obra/superpowers 100+ specialist subagents, ready to use github.com/VoltAgent/awesome… Plugins, agents and commands for Claude Code github.com/wshobson/agents Spec-driven development toolkit from GitHub github.com/github/spec-kit Batteries-included agent harness github.com/langchain-ai/deep… Graphs for agents that don't fall apart github.com/langchain-ai/lang… Connect your agents to the tools you already use github.com/modelcontextproto… Memory that survives every session github.com/thedotmack/claude… 12 principles of agents that actually work github.com/humanlayer/12-fac… Free agents course: 18 lessons github.com/microsoft/ai-agen… All of this costs $0 AI won't replace the expert. It replaces everyone who only sounds like one Bookmark it and send it to a friend still collecting generic certificates Then read the Harness, Loops, Graphs course below and build yours ↓
Your AI agent isn't lazy. It's homeless. No desk. No tools. No idea what "done" means. So you became its harness, its loop and its graph. I broke down how to stop babysitting it in 10 steps ↓
3
2
22
1,046
Holy f*ck. Anthropic CEO Dario Amodei was asked what a kid graduating high school should do now. His answer: "It will also eliminate jobs, and probably a large number of them" In 35 seconds, he lists who gets hit first: entry-level law, entry-level finance, entry-level consulting His own forecast: 50% of entry-level white-collar jobs disrupted in 1 to 5 years Every past disruption gave people time to adapt. He says this one goes deeper and moves faster Prompts → Agents → Harness → Loops → The One Running the AI His advice: learn to use AI. Just using it won't be enough. You need a harness that runs it for you Start with the repos below Every repo below is free: Anthropic's plugins for legal, finance, sales, HR and data work github.com/anthropics/knowle… Anthropic's agents for analyst work: models, memos, research notes github.com/anthropics/financ… Excel, Word, PowerPoint and PDF skills github.com/anthropics/skills Autonomous financial research agent github.com/virattt/dexter Deep research agent github.com/assafelovic/gpt-r… Full-length research reports github.com/stanford-oval/sto… 12 principles of agents that actually work github.com/humanlayer/12-fac… Batteries-included agent harness github.com/langchain-ai/deep… Claude Agent SDK github.com/anthropics/claude… Memory that survives every session github.com/thedotmack/claude… Autonomous agent loop github.com/snarktank/ralph Give your agents a browser github.com/microsoft/playwri… Claude cookbooks github.com/anthropics/claude… Free agents course: 18 lessons github.com/microsoft/ai-agen… All of this costs $0 AI won't replace you. Someone with a harness will 95% will keep applying for jobs AI already does. The 5% who build a harness will be the ones running it Save it and send it to someone just starting out Then use the Harness, Loops, Graphs course below to build yours before everyone else does ↓
1
342
Holy f*ck. Anthropic CEO Dario Amodei was asked what a kid graduating high school should do now. His answer: "It will also eliminate jobs, and probably a large number of them" In 35 seconds, he lists who gets hit first: entry-level law, entry-level finance, entry-level consulting His own forecast: 50% of entry-level white-collar jobs disrupted in 1 to 5 years Every past disruption gave people time to adapt. He says this one goes deeper and moves faster Prompts → Agents → Harness → Loops → The One Running the AI His advice: learn to use AI. Just using it won't be enough. You need a harness that runs it for you Start with the repos below Every repo below is free: Anthropic's plugins for legal, finance, sales, HR and data work github.com/anthropics/knowle… Anthropic's agents for analyst work: models, memos, research notes github.com/anthropics/financ… Excel, Word, PowerPoint and PDF skills github.com/anthropics/skills Autonomous financial research agent github.com/virattt/dexter Deep research agent github.com/assafelovic/gpt-r… Full-length research reports github.com/stanford-oval/sto… 12 principles of agents that actually work github.com/humanlayer/12-fac… Batteries-included agent harness github.com/langchain-ai/deep… Claude Agent SDK github.com/anthropics/claude… Memory that survives every session github.com/thedotmack/claude… Autonomous agent loop github.com/snarktank/ralph Give your agents a browser github.com/microsoft/playwri… Claude cookbooks github.com/anthropics/claude… Free agents course: 18 lessons github.com/microsoft/ai-agen… All of this costs $0 AI won't replace you. Someone with a harness will 95% will keep applying for jobs AI already does. The 5% who build a harness will be the ones running it Save it and send it to someone just starting out Then use the Harness, Loops, Graphs course below to build yours before everyone else does ↓
Your AI agent isn't lazy. It's homeless. No desk. No tools. No idea what "done" means. So you became its harness, its loop and its graph. I broke down how to stop babysitting it in 10 steps ↓
1
20
991