If You Want an Agent Transition, Spend Time on Initial Setup
Boris Cherny's note on automation matches what I have been telling teams for weeks: the leverage is encoding judgment into infrastructure, not burning tokens on the same bug forever.
← Index
AI Agents & Developer Tools
Deep coverage of AI agent architecture, context engineering, and developer workflows.
117 posts
What I Told Huddling Club About Two Months as a Solo Builder
Your own channel, a work harness, and documented taste matter more than raw token spend. Notes from a weekend talk with other builders.
Builders Are Already Patching Codex for Performance and Quota
Two structural issues in Codex defaults cut reasoning short and explode agent trees. Here is what I measured and the config.toml changes that fixed both.
I Mean This Bluntly: Humans Have Less Work Left in the Loop
Orchestrators absorb more execution; the scarce skill is directing systems.
Claude Code Was Billing Tens of Thousands of Hidden Tokens Per Turn
Matt Pocock's method: run /context, rank tool payloads, then trim settings.json with disable flags and bare tool-name denies so unused definitions stop riding every message.
“AI Agent” Search Is Up 23,000%. The Real Story Is Keyword Churn
Rising volume is the easy headline. Search intent is moving from build-an-agent tutorials to Gemini modes, Notion agents, Codex automation, and internal wikis.
How I Stretch Fable 5 Under Tight Rate Limits
When Fable 5 is available but capped, keep effort at high, reserve it for judgment, and push bulk work to Codex or Sonnet. Write the routing into CLAUDE.md first.
An Agent Kit for App in Toss Miniapps: Rules Before Vibes
appintoss-kit layers AGENTS.md, guides, skills, hooks, and review agents so Codex/Claude/Cursor stop missing policy, Korean UX, design rules, and store metadata.
Four Small Tools I Built to Make Daily Codex Less Annoying
A ChatGPT→Codex handoff bridge, multi-account config sharing via symlinks, a menu-bar status app, and an imagegen API with provider fallback.
Stop Installing Skills Globally. Symlink Them Per Project
A full global skills library loads junk into every session. Keep one warehouse of clones and symlink only what each repo needs.
Twelve Codex Tricks Power Users Quietly Share
ChatGPT-as-MCP planning, multi-account gateways, Excalidraw intent, imagegen fan-out, remote handoff, grill-me PRDs, kanban agents, and multi-tool control.
The “Senior Who Turns 50 Lines Into One” Is Now an Agent
Encode review taste as rules and examples, then let an agent apply the same compressions a strong reviewer would.
사용자의 반복 작업을 스스로 skill 로 만드는 프롬프트가 큰 화제입니다
Blog version of a 2026-05-30 LinkedIn note — same claims, cleaned structure for reading.
"진짜" AI 네이티브 조직을 "현실적으로" 구축하려면 어떻게 해야할까요
Blog version of a 2026-05-29 LinkedIn note — same claims, cleaned structure for reading.
MCP 프로토콜이 7월 28일에 완전히 바뀝니다
Blog version of a 2026-05-26 LinkedIn note — same claims, cleaned structure for reading.
핸드폰으로 에이전트를 원격 제어하는 방법을 총정리했습니다
Blog version of a 2026-05-25 LinkedIn note — same claims, cleaned structure for reading.
디자이너라면 한 번 읽어보시면 좋을 것 같아요
Blog version of a 2026-05-25 LinkedIn note — same claims, cleaned structure for reading.
에이전틱 코딩으로 생산성을 올리고 싶다면, 이 5가지 개념부터 잡아야 합니다
Blog version of a 2026-05-23 LinkedIn note — same claims, cleaned structure for reading.
AI 에게 맥락을 제대로 주고 싶다면, 관계를 Graph 로 만드세요
Blog version of a 2026-05-21 LinkedIn note — same claims, cleaned structure for reading.
Codex 제대로 쓰는 법, 9가지 요소로 정리한 이 글 하나면 충분합니다
Blog version of a 2026-05-21 LinkedIn note — same claims, cleaned structure for reading.
대한민국 디지털 정부 디자인 시스템 KRDS, AI Skill 이 없길래 직접 만들었습니다
Blog version of a 2026-05-19 LinkedIn note — same claims, cleaned structure for reading.
구글 I/O 2026 은 충격적인 업데이트의 연속이었습니다
Blog version of a 2026-05-19 LinkedIn note — same claims, cleaned structure for reading.
4B 모델 87점 vs 14B 모델 75점. 차이는 하네스입니다
Blog version of a 2026-05-19 LinkedIn note — same claims, cleaned structure for reading.
Claude Code Codex
Blog version of a 2026-05-16 LinkedIn note — same claims, cleaned structure for reading.
Codex 도 핸드폰 안으로 들어왔습니다
Blog version of a 2026-05-15 LinkedIn note — same claims, cleaned structure for reading.
Codex 로 PPT 예쁘게 만들기
Blog version of a 2026-05-13 LinkedIn note — same claims, cleaned structure for reading.
요즘 한국 AI 씬에서 가장 진솔하면서도 유익한 콘텐츠를 만든다고 생각하는 두 분과 각각 커피챗을 가졌다. 한 분은 "까칠한 AI"라는 유튜브를 운영하시는 SpaceY 황현태 대표
Blog version of a 2026-05-12 LinkedIn note — same claims, cleaned structure for reading.
오늘부터 하루에 하나씩 저의 업무 전용 AI 도구를 무료로 공개합니다.
Blog version of a 2026-05-12 LinkedIn note — same claims, cleaned structure for reading.
제가 좋아하는 AI 디자인 관련 레퍼런스를 모두 모았습니다
Blog version of a 2026-05-10 LinkedIn note — same claims, cleaned structure for reading.
Google 과 함께 코엑스에서 발표하고 왔습니다
Blog version of a 2026-05-07 LinkedIn note — same claims, cleaned structure for reading.
디자인 킬러 DESIGN.md 를 가장 잘 사용하는 워크플로우를 정리했습니다.
Blog version of a 2026-05-05 LinkedIn note — same claims, cleaned structure for reading.
Codex 가 무거워지기 시작했다면 이 글을 한 번 보세요.
Blog version of a 2026-05-04 LinkedIn note — same claims, cleaned structure for reading.
AI를 잘 쓰는 사람과 못 쓰는 사람의 차이, 저는 읽는 능력 하나라고 봅니다
Blog version of a 2026-05-02 LinkedIn note — same claims, cleaned structure for reading.
한국어로 Claude를 쓰면 영어보다 2.59배 비쌉니다
Blog version of a 2026-04-29 LinkedIn note — same claims, cleaned structure for reading.
Anthropic 이 클로드 코드의 문제를 공식적으로 인정했습니다
Blog version of a 2026-04-23 LinkedIn note — same claims, cleaned structure for reading.
슬라이드, 이제 바로 영상이 됩니다.
Blog version of a 2026-04-21 LinkedIn note — same claims, cleaned structure for reading.
Codex computer use
Blog version of a 2026-04-19 LinkedIn note — same claims, cleaned structure for reading.
AI 토큰 소모를 60% 이상 절약하는 무료 방법 10가지를 모았습니다
Blog version of a 2026-04-18 LinkedIn note — same claims, cleaned structure for reading.
병렬 에이전트를 잘 관리하려면 격리하고 기록하고 합치는 구조가 먼저입니다
Blog version of a 2026-04-13 LinkedIn note — same claims, cleaned structure for reading.
프론티어 기업은 AI 도입에 전략이 없었습니다
Blog version of a 2026-04-10 LinkedIn note — same claims, cleaned structure for reading.
오늘부터 Codex Pro 플랜 사용자는 2배 이벤트를 다시 시작합니다
Blog version of a 2026-04-10 LinkedIn note — same claims, cleaned structure for reading.
Gemini NotebookLM
Blog version of a 2026-04-09 LinkedIn note — same claims, cleaned structure for reading.
공식적으로 Claude 의 뇌와 손이 분리되기 시작했습니다
Blog version of a 2026-04-09 LinkedIn note — same claims, cleaned structure for reading.
AI 콘텐츠 수집 아직도 수동으로 하고 계시면, 이 5개부터 당장 깔아보세요
Blog version of a 2026-04-08 LinkedIn note — same claims, cleaned structure for reading.
4월 들어서 에이전트 개발에 완전히 올인하고 있습니다.
Blog version of a 2026-04-06 LinkedIn note — same claims, cleaned structure for reading.
Eight Hooks That Guarantee AI Agent Reliability
CLAUDE.md rules get followed about 80% of the time. Hooks get followed 100% of the time. After six months of testing, these are the eight I never removed.
Claude Code in 2026: Layers Matter More Than Tools
I installed three popular Claude Code extensions and productivity barely moved. The problem was never which tools to pick.
Why Your Codex Config Isn't Working: The .codex/ Folder Problem
I edited config.toml, wrote rules in AGENTS.md, and nothing stuck. Turns out the folder structure itself was the issue, not my settings.
Codex Moved Into Claude Code, and That Says Everything
OpenAI shipped Codex as a Claude Code plugin on the same day Anthropic announced Computer Use. I think it's the smartest concession of 2026.
2026년 Claude Code 의 승부처는 도구가 아니라 레이어입니다
Blog version of a 2026-03-29 LinkedIn note — same claims, cleaned structure for reading.
️ 2026년 1분기 AI 활용 4가지 방향성이 드러나기 시작했습니다
Blog version of a 2026-03-28 LinkedIn note — same claims, cleaned structure for reading.
Claude Code Channels Changed How I Work Away From the Terminal
A month ago I couldn't leave my laptop during a build. Three features in four weeks fixed that.
The 10-Hour Skill Beats the 10-Minute Skill Every Time
I thought a single SKILL.md file was enough. Then I saw how Anthropic's own team structures theirs, and rebuilt everything.
Four Contexts That Decide Whether AI Helps or Wastes Your Time
I spent a weekend stuffing 100MB of PDFs into an agent. Performance got worse. Mapping what I was feeding into four categories finally showed me why.
12 Free Skills That Fix AI Slop Design
I tested dozens of design skills for AI coding agents. Most didn't last a week. These 12 are the ones I still use.
Create Three Spec Files Before Using Claude Code and Codex
I spent a year getting wildly inconsistent results from Claude Code and Codex. Three spec files, each with a distinct role, fixed it.
After AI Agents Write the Code, the Human's Job Is Visualization
Agents writing code is just the start. To review PRs and explain architecture to teammates, you need visualization tools.
5 Settings That Separate the Top 0.01% Claude Code and Codex Users
Subscribing puts you in the top 0.3%. These five configurations — agents, teams, MCP, monitoring, automation — push you into the top 0.01%.
31 AI Coding Agent Terms You Should Know, Sorted Into Five Pillars
I classified every term I kept encountering while using Claude Code and Codex daily. Five groups emerged, and they map the entire system these tools run on.
Claude Code 29 Tools vs Codex 7 Tools: The Design Philosophies Are Polar Opposites
I dug into SDK type definitions and system prompts for both tools. The 29 vs 7 gap isn't about feature count. It's about two fundamentally different answers to the same question: how should an AI coding agent interact with your system?
AI 에이전트에게 작업을 시킨 다음 인간이 해야 할 일은 시각화입니다
Blog version of a 2026-03-12 LinkedIn note — same claims, cleaned structure for reading.
Create Three Spec Files Before Giving Work to Claude Code or Codex
After a year of agent-assisted development, I found that structured spec files fixed the inconsistency problem better than any prompt technique.
How Codex Solves the Compaction Problem Differently
I reverse-engineered how Codex handles context overflow compared to Claude Code. The answer involves AES encryption, session handover patterns, and KV cache tricks.
Shopify's CEO Built a Search Engine That Revives 700 Claude Code Sessions in One Second
Shopify CEO Tobias built QMD, an open-source search engine. Connect it to Claude Code and every session gets persistent memory.
4 Tool Design Principles Claude Code Learned After 3 Rebuilds
Anthropic's Claude Code team rebuilt their tools three times. Fewer tools made the AI perform better. Here are four hard-won design principles.
Claude Code Sub-Agents Save 25x Tokens in Your Main Session
Your AI isn't getting dumber. Your main session is overloaded. Sub-agents keep it lean and accurate for over an hour.
Claude Code Just Reset Everyone's Weekly Limits to Zero — Here's What Happened
A race condition between Auto Memory and context compaction in Claude Code v2.1.59–v2.1.61 broke prompt caching and corrupted sessions. Anthropic reset all weekly limits as compensation.
Two Tools Every Claude Code User Needs - Agentation and Readout
Agentation gives AI agents pixel-perfect visual feedback via CSS selectors. Readout replays Claude Code sessions like video. Together they eliminate the two biggest friction points in AI-assisted frontend development.
I Dug Through 300 Agent Failure Logs. The Problem Was Never the Prompt.
An open-source context engineering skillset just crossed 10k GitHub stars. After applying it to my own agent stack, I finally understand why agents fail.
My Agent Called a Failed API 5 Times—The Bug Wasn't in the Code
When an agent repeats the same failing API call, code review won't help. Traces are the new source code for debugging AI agents.
I Was Too Lazy to Write CLAUDE.md — Turns Out That Was the Right Call
New benchmark data shows AGENTS.md and CLAUDE.md context files actually hurt coding agent performance. Sometimes laziness is the best engineering decision.
4 Updates from Anthropic, OpenAI, and Google: The Hidden Pattern
Three companies updated their coding agents at the same time. The directions overlap. The real battleground isn't models; it's how fast they absorb developer workflows.
5 AI Predictions for 2026: Agents, Chips, and Historic Exits
From the SaaSpocalypse to model-specific silicon, five bold predictions for where AI is heading in 2026, with roughly 50% confidence of getting them right.
The Cache Design That Cuts Claude Code API Costs by 90%
My API costs jumped 10x when the cache broke in production. The same day, Anthropic engineers explained exactly why.
From 6.7% to 68.3% Task Success: The Harness Made the 10x Difference, Not the Model
What LangChain's Terminal Bench results and the hashline format experiment revealed. The same model flipped leaderboard rankings, and the reasons came down to three things: prompts, tools, and middleware.
The Web's Top Customer Is Shifting from Humans to AI
From Cloudflare and Vercel's Markdown for Agents to Google's WebMCP, reading and writing are being standardized simultaneously, ushering in the Agent-Native Web era.
5 SKILL.md Body Writing Rules Hidden in Anthropic's Docs
Five SKILL.md body writing principles buried in Anthropic's official documentation. From separating description and body roles to embedding verification loops.
Why OpenAI Hired the OpenClaw Creator - The AI-Native Messenger Era Begins
Peter Steinberger joining OpenAI isn't just a talent grab. It signals the dawn of AI-native messengers that could redefine how we communicate.
How OpenAI Built 1 Million Lines of Code Using Only Agents: 5 Harness Engineering Principles
OpenAI's Codex team built a 1M-line codebase using only AI agents. Here are the five harness engineering principles they discovered along the way.
Claude Code Agent Teams - From Setup to Shortcuts
A practical guide to Claude Code's new multi-agent teams feature: activation, keyboard shortcuts, terminal compatibility, task management, and known limitations.
OpenAI & Google's 2026 Declaration: The AI Price War Has Begun, and Now Is the Best Time to Jump In
OpenAI and Google are racing to launch affordable AI plans while Chinese competitors shatter price floors. Here's why this moment is your best entry point.
The AI Wrapper Era Is Over. The Age of Claude Agent Wrappers Has Begun.
Anthropic's Tariq Shihipar breaks down what it actually takes to build production-grade agents - from Bash-first tooling to file-system-driven context engineering.
Claude Code Was for Developers - Cowork Is for Everyone
Anthropic launches Cowork, an autonomous agent that reads, edits, and creates files on your local machine. Vibe coding meets vibe working.
ChatGPT Is Android, Claude Is iOS
Why $300B evaporated from SaaS stocks as ChatGPT and Claude race to become the AI app store - and what the 2008 mobile wars tell us about what comes next.
How the Creator of Claude Code Actually Uses Claude Code
Boris Cherny's workflow hit 5K likes in 2 hours. His setup is simpler than you'd expect - parallel sessions, plan mode, CLAUDE.md, and verification loops.
The Claude Code Setup Guide That Made Me Rethink Everything
An Anthropic hackathon winner's 10-month Claude Code configuration - context management, hooks, subagents, and the principles that actually matter.
Claude Code Skills Setup: 1 Person Doing the Work of 30
Six Claude Code skill combinations that let a small team run a full-stack business - from marketing and video to UI design and code quality.
4 Claude Code Skills That Actually Survived Out of 100K
After installing hundreds of AI coding agent skills, only 4 made it into my daily workflow. Here's what survived the weekend audit.
What Claude Code's Task System Reveals About the AI-Native Engineer
Claude Code renamed Todo to Task. It looks like a small change, but it marks the beginning of a completely different system - one built for AI swarms.
Claude HUD: The Plugin That Made AI Agents Observable
A game-style status bar for Claude Code that shows context usage, active tools, sub-agents, and todo progress in real time.
The Weekend Clawdbot Sold Out the Mac Mini - and Why Kakao Should Be Worried
Clawdbot proved that AI agents running locally on your own hardware can replace messenger apps. Here's why that threatens every chat platform.
Context7: Use Skills and Subagents, Not Just MCP
Connecting Context7 via MCP floods your main context with docs. Skills and subagents isolate queries, keeping long coding sessions stable.
Every App Becomes an API in the Age of Personal Agents
Why YC and OpenClaw leaders believe software is being rebuilt for agents - and what it means for developers building products right now.
Everything Is Becoming a Skill: AI Is Now the One Using Your Service
With AI reading 50% of developer docs and bot traffic outpacing humans 3-to-1, services are racing to package their knowledge as agent skills. Here's what's driving the shift.
Karpathy's Wake-Up Call: The New Abstraction Layer Every Developer Must Learn
Andrej Karpathy admits he's never felt this behind as a developer. Here's the new AI agent abstraction layer he says you must master - or risk falling 10x behind.
Meta's $2.5B Manus Acquisition - Its Core Tech Is Now Open Source
The file-based memory system behind Manus's $2.5 billion valuation is now a free Claude Code skill. Here's why it matters for every AI agent builder.
Manus Acquired by Meta for $300M Reveals Core Agent Development Principles with LangChain
Manus shared the hard-won lessons behind building production AI agents - from context rot to evaluation rethinking - in a joint presentation with LangChain.
Inside the $3.6B Secret Behind Manus: Why AI Agents Actually Fail
Meta acquired Manus for $3.6 billion. The secret wasn't a bigger model - it was context engineering. Here's what most AI agents get wrong.
Multi-Agent Architecture: Splitting Blindly Will Backfire
Not all multi-agent patterns are equal. Learn when subagents, skills, handoffs, and routers actually outperform a single agent - with real scenarios and numbers.
The One Article That Unblocked My Multi-Agent Architecture
Orchestration patterns, communication methods, memory management, and production pitfalls - a practical breakdown of everything I struggled with when designing multi-agent systems.
Dissecting Oh-My-OpenCode and the Future of Context Engineering
A deep dive into Oh-My-OpenCode's multi-agent orchestration architecture - how programmatic context isolation, parallel execution, and evidence-based research are redefining what AI coding agents can do.
7 Building Blocks of AI Agent Design, Learned from Opencode Docs
Opencode's open-source documentation doubles as an introductory guide to agent architecture. Here are the seven core concepts every developer should understand.
The Era of AI Deploying Production Code Overnight
How a Claude Code plugin named after Ralph Wiggum is redefining autonomous coding through iterative loops, memory architecture, and stop hooks.
6 AI Agent Trends Established Worldwide in January 2026
Six battle-tested AI agent patterns that emerged globally in one month - from persistent loops to multi-agent orchestration.
6 Principles of AI Agent Development Established Globally in One Week
Context engineering took the world by storm in early 2026. Here are six battle-tested principles from Manus, Cursor, and Claude Code that define modern AI agent development.
10 Principles of AI Coding from the OpenClaw Creator
Peter Steinberger, who built GitHub's fastest-starred project, shares 10 hard-won principles for working with AI coding agents.
Why Claude Code Dropped Todos and Slash Commands
Anthropic replaced TodoWrite with Tasks and Slash Commands with Skills in two days. Both changes point in the same direction - unhobbling the model.
Will UI Really Disappear in the AI Era of 2026?
Claude Code and AI avatar apps prove users want results, not complex interfaces. Here is what actually disappears when UI gets abstracted away, and what remains.
My Terminal Setup for AI Agentic Coding
How I use Ghostty, Yazi, Fish, and LazyGit to run multiple AI agents in parallel - a lightweight terminal stack built for agentic workflows.
Hacking Day UI UX QA Flask Folo
Blog version of a 2026-02-04 LinkedIn note — same claims, cleaned structure for reading.
Claude Code AskUserQuestion
Blog version of a 2026-02-02 LinkedIn note — same claims, cleaned structure for reading.
2026년 1월, AI 한테 UI 센스 심어주는 도구 3가지
Blog version of a 2026-01-18 LinkedIn note — same claims, cleaned structure for reading.
Context7, 단순 MCP 연결 말고 스킬, 서브에이전트로 컨텍스트를 절약하세요
Blog version of a 2026-01-17 LinkedIn note — same claims, cleaned structure for reading.
MCP Tool Search Tool
Blog version of a 2026-01-15 LinkedIn note — same claims, cleaned structure for reading.
Cursor Codex Gemini Cline AI Plan Mode
Blog version of a 2026-01-13 LinkedIn note — same claims, cleaned structure for reading.
바이브 코딩 시대, 창업가의 역할이 '코더'에서 '편집장'으로 바뀌고 있습니다.
Blog version of a 2026-01-11 LinkedIn note — same claims, cleaned structure for reading.
AI 슬롭을 막는 컨텍스트 엔지니어링 관점에서의 클로드 코드 실전 가이드
Blog version of a 2026-01-08 LinkedIn note — same claims, cleaned structure for reading.