Claude Code

Agentjacking: how a fake bug report hijacks your coding agent

Agentjacking is an attack disclosed in June 2026: an attacker plants a fake error report, your AI coding agent reads it as instructions, and runs their code with your credentials. The nasty part is that telling the agent to ignore untrusted input doesn't stop it, and your security tools see nothing wrong.

Agentjacking：一封假錯誤報告，就能讓 coding agent 替駭客跑指令

Agentjacking 是 2026 年 6 月揭露的攻擊：攻擊者塞一封假的錯誤報告，AI coding agent 把它讀成指令、用你的權限跑了攻擊者的 code。最麻煩的是，在系統提示裡叫 agent 別理它也擋不住，資安監控也完全看不出異常。

AI 寫 code 為什麼又搬回終端機了

這兩年 AI 寫 code 的重心，悄悄從嵌在 IDE 裡的補全，搬回了終端機（Claude Code、Codex CLI 那一類）。我覺得這不是復古，是因為 agent 變成一個「你不盯著看的 process」，而終端機本來就是為這種東西設計的。

Why coding agents are moving back to the terminal

The coding agents developers reach for now — Claude Code, Codex CLI, Aider — are terminal programs, not IDE plugins. An IDE is built around a human at the keyboard, the terminal around processes you don't babysit. When AI coding became a job instead of a keystroke, it moved home.

When an AI says "done," ask it to show you

An AI's 'done' sounds the same whether the work happened or not. The fix is one small habit: don't take its word for it, ask it to show you a result you can check yourself, sized to the task.

AI 說「完成了」，怎麼確認它真的做完？

AI 回報「完成了」的時候，真的做完、做一半繞過去、方向整個誤會，那段話讀起來幾乎一樣。與其判斷那句話可不可信，不如養成一個反射：給我看一個我自己查得到的東西，commit、測試輸出、diff。

Claude Code

See all 20

When an AI says "done," ask it to show you

An AI's 'done' sounds the same whether the work happened or not. The fix is one small habit: don't take its word for it, ask it to show you a result you can check yourself, sized to the task.

AI 說「完成了」，怎麼確認它真的做完？

Agentic OS

See all 11

When an AI says "done," ask it to show you

An AI's 'done' sounds the same whether the work happened or not. The fix is one small habit: don't take its word for it, ask it to show you a result you can check yourself, sized to the task.

AI 說「完成了」，怎麼確認它真的做完？

Claude Code 多了個 dynamic workflows，我打開那段 JS 看了一下

Claude Code 5/28 釋出 dynamic workflows，跟 Opus 4.8 同一天上。比起「能開 1000 個 subagent」那個數字，更關鍵的是 orchestration 那段 JS 是 Claude 寫的、不是 Claude 在跑——這件事其實滿值得想一下的。

How Claude Code's Dynamic Workflows Run 1,000 Subagents

Claude Code's new dynamic workflows hand the orchestration plan over to a JavaScript script that Claude writes. The runtime executes it with up to 1,000 subagents — 16 concurrent — and Claude's context only sees the final cross-checked answer.

怎麼寫你的第一個 skill — 從一個煩躁的 prompt 開始

你的第一個 skill 不會長得像書裡那些 production-grade 的成熟形態，它會長得像「你重複打三次的同一個 prompt」。從那裡開始，比從一個成熟框架的 skill 倒著學容易很多。

13 行的 skill：AI 起稿，我事後才看懂

我請 AI 幫我寫一個能從 Claude Code 呼叫 Codex CLI 的 skill，它給我 13 行 markdown。13 行很小，但 skill 跟 prompt 真正的差別不在這 13 行裡——在它指過去的那一份東西裡。

Architecture

See all 9

AI 寫 code 為什麼又搬回終端機了

Why coding agents are moving back to the terminal

Claude Code 多了個 dynamic workflows，我打開那段 JS 看了一下

How Claude Code's Dynamic Workflows Run 1,000 Subagents

The Skill Your Annoyed Prompt Becomes

Your first Claude Code skill won't look like the polished examples in tutorials. It'll look like a prompt you've typed three times in a row, saved into a four-line markdown file. This post walks that minimum shape, shows the three things that break, and compares it to a real seventeen-line production-grade skill from the framework I use daily.

怎麼寫你的第一個 skill — 從一個煩躁的 prompt 開始

Governance

See all 9

When an AI says "done," ask it to show you

An AI's 'done' sounds the same whether the work happened or not. The fix is one small habit: don't take its word for it, ask it to show you a result you can check yourself, sized to the task.

AI 說「完成了」，怎麼確認它真的做完？

Claude Fable 5: First Public Mythos-Class Model, One Day In

Anthropic released Claude Fable 5 on June 9 — the first publicly available Mythos-class model, one tier above Opus. What it is, what it costs, the June 22 deadline on the subscription window, and what changed when I pointed three real projects at it for a day.

Claude Fable 5 是什麼？第一個公開的 Mythos 級模型，加上我第一天的使用心得

Anthropic 6/9 釋出第一個公開的 Mythos 級模型 Claude Fable 5。這篇整理它跟 Opus 4.8 的關係、定價、6/22 截止的訂閱免費期，加上第一天把三個專案丟給它跑的心得：它對治理流程的遵守程度是真的，token 也是真的兇。

No evidence, no completion

No evidence, no completion: the one rule that closes most AI agent failures. A task isn't done until it produces a verifiable artifact (commit SHA, test output).

Work Log：跨 session 的記憶機制

AI 代理每個新對話都失憶？Work Log 用一份 markdown 記錄任務進度與決策，讓 Claude Code 跨 session 接續，不用每次重講背景。

LLM

Claude Fable 5: First Public Mythos-Class Model, One Day In

Claude Fable 5 是什麼？第一個公開的 Mythos 級模型，加上我第一天的使用心得

Claude Code 多了個 dynamic workflows，我打開那段 JS 看了一下

How Claude Code's Dynamic Workflows Run 1,000 Subagents

The Skill Your Annoyed Prompt Becomes

What a 13-Line Skill Leaves Out

I asked Claude to draft me a skill that calls OpenAI's Codex CLI. It came back as thirteen lines of markdown. The thirteen lines aren't the skill — they point to where the skill actually lives. That split between dispatcher and contract is what separates a skill from a prompt.

Skills

Prompt Engineering