Always in the middle of something.

從一段重打到第三次、懶得再打的 prompt，到一套會自己留收據的系統，中間隔了好幾層。這篇一層一層走一遍：每一層都是被前一層某個具體的毛病逼出來的，而這條路這個部落格自己走過，每一層都留了一篇當時的記錄。

A browser-based JSON formatter that prettifies, minifies, and points to the exact line and column where a payload won't parse — all in your browser. Plus the cases where jq or python -m json.tool is all you need.

A prompt and a skill can contain the same words. The difference is the machinery around them: a skill is a file the model loads on its own when your request matches its one-line description, and it can declare inputs, tools, and a scope it won't cross. Here's how that loading works and why it lets you build things a prompt can't.

OpenAI shipped GPT-5.6 as Sol, Terra, and Luna on July 9, 2026, and quietly ended the dedicated Codex checkpoint. What replaced it is a reasoning-effort dial whose top notch spawns subagents. This post lays out the three models, the six effort levels, and what the independent benchmarks measured.

GPT-5.6 在 2026 年 7 月 9 日全面上線，分成 Sol、Terra、Luna 三階，而這一代的 Codex 沒有專用模型。專用 checkpoint 換成一格叫 ultra 的 reasoning effort，官方說它會叫 subagent 出來把大任務拆開做。這篇整理三顆模型怎麼分、六段 reasoning effort 是什麼，以及幾份獨立評測量到的數字。

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 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。

打「貓」「小貓」「喵星人」，AI 怎麼知道講的是同一種東西？答案是 embedding：把文字變成座標，再用夾角量意思相不相近。這篇拆開它怎麼運作，順便聊聊「king − man + woman = queen」那個經典其實灌了點水。

Search for 'make my laptop quieter' and get a page about 'reducing fan noise' with zero words in common. That's embeddings: text turned into coordinates, with meaning measured as the angle between two points. Here's how the trick works, and where it's oversold.

Token counts land on your API bill and decide whether a prompt fits the context window. Here's why Chinese usually costs more tokens than English, why eyeballing it fails, and how to see the real number for any chunk of text.