Tag: claude-code
All the articles with the tag "claude-code".
-
Claude Seeing Ghosts Four Nights Straight: A Log of Opus 4.8 Fabricating Tool Output
-
Claude 半夜見鬼連續四天：Opus 4.8 捏造工具輸出實錄
-
Opus 4.8 Cries 'Prompt Injection,' Codex GPT-5.5 Tracks Down the Real Cause: a Worktree Race
-
Opus 4.8 搞烏龍報『提示注入攻擊』，Codex GPT-5.5 揪出 worktree race 真因
-
Running Taiwan Sports-Lottery Odds With Mobile Claude Code, Standing in the Lottery Shop: the Odds Are Garbage, You'll Lose
-
在彩券行用手機 Claude Code 算運彩賠率：台運賠率爛透了，誰下誰賠
-
Two Mindsets for Getting Unstuck: Logical Thinking, and the AI Pedagogy of Never Making the Same Mistake Twice
-
卡關時的兩個心態：用邏輯思考，與『不貳過』的 AI 教育學
-
Teaching a Talent Agency to Use Claude Code, and the Soft Landing of the DK Confidence Curve
-
帶藝人經紀公司用 Claude Code，與 DK 信心曲線的軟著陸
-
The Perpetual Token-Burning Machine: Anthropic Really Is Trying
-
燒 token 永動機：Anthropic 真的很努力
-
The Five-Step SOP for Sneaking Claude Code Runs From the Hotel Over Dinner
-
出門吃飯前，讓 Claude Code 在旅館背景偷跑的五步 SOP
-
From Miasma to Hades: How One Group Turned AI Tools Into a Supply-Chain Attack Vector
-
從 Miasma 到 Hades：同一個攻擊組織如何把 AI 工具當成供應鏈攻擊媒介
-
A Few Takeaways from the Claude Code Anniversary Talk
-
Don't Dump the Last Session's JSON — Check Your Harness First
-
Claude Code 週年對話的幾個重點
-
別把上一個對話的 JSON 餵給它，先體檢 harness
-
Worried About Agents Going Rogue? Start Isolated, Hand Off Gradually
-
A Few Months With Claude Code: The Companion Tools and Resources I've Accumulated
-
Claude Code To-Do Tiering: Auto / Review / Collaborate / Manual
-
The Day Opus 4.8 Went Haywire: tool call cannot be parsed
-
I Built a "Tinghao SKILL" — Distilling a Finance YouTuber's Perspective Into AI
-
擔心 agent 失控？先在隔離環境逐步放手
-
用 Claude Code 這幾個月，累積下來的周邊工具與學習資源
-
Claude Code 待辦分級制：自動／審核／協作／手動四級分流
-
Opus 4.8 全線抽風的一天：tool call cannot be parsed
-
我做了一個「庭皓 SKILL」——把財經 YouTuber 的觀點蒸餾成 AI
-
The Blind Spot in AI Self-Verification — Why You Need an Uninformed Agent to Audit
-
AI 自我驗證的盲點——為什麼要請一個不知情的 agent 來稽核
-
Better Context Beats a Stronger Model — repowise's Five-Layer Architecture
-
給 LLM 更好的 context，勝過換更強的模型——repowise 的五層架構
-
A Student's Seven-Agent Setup That Wouldn't Run — A Cautionary Tale About Harness First
-
學員自建七個 agent 跑不動——一個 harness 為重的反面教材
-
Leveling Up the iPad Workflow — Surviving Internet and Power Outages, and Why I Ditched the Magic Keyboard
-
iPad 工作流進階——斷網斷電的備案，與我為什麼放棄妙控鍵盤
-
Love-Hate With Anthropic — Sneering at the Elitism While Devouring Their Engineering Blog
-
又愛又恨 Anthropic——一邊嫌精英心態，一邊狂讀他們的技術部落格
-
Racing for the Fastest Summary — Opus 4.8's 244-Page System Card, and How I Read It With 20 Agents in Half an Hour
-
拼全網最速——Opus 4.8 系統卡 244 頁重點，外加我怎麼用 20 個 agent 半小時讀完
-
The More Rules You Add, the Less Claude Listens — I Sent a Team of Agents to Trim My Setup and Cut 36% of Always-On Context
Late at night I got Opus 4.8, and my weekly quota happened to reset. The first thing I did was put my harness on a diet. A COO student who had burned through 88% of his 1M context made me face one thing squarely: the more rules you add, the less the model listens.
-
規則越加，Claude 越不聽話——派一隊 AI 重整設定，常態上下文省 36%
半夜拿到 Opus 4.8、額度剛好重置，我做的第一件事是減肥我的 harness。一個 COO 學員把 1M 上下文用到 88% 的案例，讓我認真面對一件事：規則加越多，模型反而越不聽話。
-
Claude Code Ecosystem Overview — Skills, Hooks, Agents in Full Bloom
-
Claude Code 生態系盤點——Skill、Hook、Agent 全面開花
-
Running Claude Code on iPad — tmux + Tailscale + Moshi Setup Guide
-
iPad 跑 Claude Code 全攻略——tmux + Tailscale + Moshi
-
Cross-Model Review — Stop Letting AI Grade Its Own Homework
-
跨模型互審——讓 AI 不再自己審自己
-
Have Claude scan an MCP before you install it—it found 7 vulnerabilities and still said it's safe to install
-
裝 MCP 之前先讓 Claude 幫你掃——找到 7 個漏洞，但說這個可以裝
-
Three moves to firepower-demo Agentic AI at your company—desktop cleanup, conflicting meeting notes, multi-version emails
-
在公司火力展示 Agentic AI 的三招——桌面整理、會議紀錄、信件多版本
-
Covey's time management matrix, now with Claude Code—how each quadrant shifts
-
Covey 七個習慣的時間矩陣放上 Claude Code 之後——每一格的時間分配怎麼變
-
Two MCP servers for Taiwan public data—what mcp-taiwan-legal-db and Twinkle Hub are each good for
-
台灣公開資料兩個 MCP 評測——mcp-taiwan-legal-db 跟 Twinkle Hub 各適合什麼場景
-
How to tune Opus 4.7's effort, and why I still keep non-coding tasks on Sonnet 4.6 medium
-
Opus 4.7 的 effort 怎麼調，跟為什麼我非編程任務還是停在 Sonnet 4.6 medium
-
Claude Code Desktop vs CLI right now—why I think beginners should just start with the desktop app
-
Claude Code 桌面版跟終端機現在差在哪——我為什麼覺得初學者直接從桌面版開始就好
-
Judging AI Risk on Two Axes: Reversibility × Environment Isolation
-
判斷 AI 風險的兩個維度：可逆性 × 環境隔離
-
Moving from Claude Code to Codex: I Built a Skill to Handle It
-
從 Claude Code 搬家到 Codex：我做了一個 Skill 自動處理
-
Is your AI subscription worth it? A night market steak framework
-
AI 訂閱費值不值？夜市牛排思考框架
-
Teaching non-technical people AI: five things the tech world assumes everyone knows
-
帶麻瓜學 AI：技術圈習以為常、但他們從來不知道的五件事
-
AI community digest W18: Claude detects Codex cheating, GPT 5.5 guardrails confuse everyone
-
AI 社群週報 W18：Claude 偵測到你在偷用 Codex、GPT 5.5 的過濾器到底鬆了沒
-
1M context isn't about holding more — it's about managing better
-
Knowledge base deep clean: using AI to find months-old problems
-
1M Context 不是裝更多，而是管理更好：Claude Code Session Management 心法
-
Claude Code this week: 1M context as a management problem, quota anxiety goes collective, SpaceX takes Cursor
-
Two Worlds of Web-Based Claude Code Setup
-
See the Outside, Think the Inside: Two Claude Skills That Pair
-
Claude Code 這週整理：1M Context 怎麼用、額度焦慮集體發作、Cursor 被 SpaceX 搶走
-
網頁版 Claude Code 環境設定的兩個世界
-
看清外面、想清裡面：兩個 Claude Skill 的分工
-
知識庫大掃除：用 AI 幫你找到埋了幾個月的爛帳
-
A Legal Counsel Used Claude Code to Catch Patent Infringement — Days Down to Hours
-
法務用 Claude Code 抓專利侵權，從幾天變成一天
-
My Claude Code Automation Stack: /loop, Monitor Tool, the Skill System, Obsidian Reflection Reviews
-
我家的 Claude Code 自動化工作流：/loop、Monitor Tool、skill 體系、Obsidian 反思週報
-
Six Months with Claude Code: From First Affection to Hesitation, From Overtime to Detox, From Mandarin Circles to the Thai Community
-
Claude Code 用了半年：從好感到猶豫、從加班到 detox、從中文圈到泰國社群
-
Two Months of Claude Code Harness Ops: Crashes, Zombies, Cache Misses, and Plan Mode Edges
-
Claude Code Harness 兩個月運維史：當機、殭屍、快取、Plan Mode 的邊邊角角
-
Claude Code This Week: Offloading CLAUDE.md, Quota-Saving Tricks, Ghostty Click Fix
-
Claude Code 這週整理：分擔 CLAUDE.md 的進階設定、省 quota 技巧、Ghostty 點擊設定
-
Opus 4.7 After One Week: From the System Card to the Roasting, Where Did It Go Wrong?
-
Opus 4.7 上線一週：從系統卡到火烤文，到底哪邊出了問題
-
90% of People Are Two Years Behind: The Real AI Adoption Gap
-
Claude Code Subscription Guide: The Real Gap Between Pro and Max
-
Why Claude Code Only Reaches Its Full Potential in the Terminal
-
90% 的人落後兩年：AI 採用的真實落差
-
Claude Code 訂閱怎麼選：從 Pro 到 Max 的真實落差
-
為什麼 Claude Code 在終端機裡才是完全體