Hands-on prompt engineering and LLM evaluation. Built around a diagnose-then-ship loop: forensic transcript critique ▎ that names the exact failure mode, paired with a fixer that produces complete, deployment-ready rewrites — not ▎ vague advice.
Turn "I feel uncomfortable but do not know how to say it" into a boundary plan you can actually keep. This download skill helps define the boundary type, violation signals, acceptable and unacceptable behavior, a clear communication script, pushback response, executable consequence, and support needed. Built to prevent silent endurance, guilt-based yeses, repeated family / partner / coworker overreach, or refusal that swings between explosion and collapse.
Get your pitch scored by 3 experts — an investor, a buyer, and a skeptic. Find weak points before your audience does.
你跟 AI 說「幫我規劃一個 X」，它兩秒就吐出一份看起來無懈可擊的計畫——問題是，裡面一半假設是它自己掰的，你分不出是哪一半。照著做，兩天後才發現方向從第一步就錯了，全部重來。這支反過來做：先把你腦中真正在想的東西窮追猛打地挖出來，逼你自己想清楚，在你確認達成共識之前，它一個字都不動。四條硬規則讓它不惹人厭：每題都附推薦答案不丟空白問卷、事實自己查只問真正該你決定的、一次只問一題不讓你腦當機、沒共識就不開工。附 18 個追問鏡頭——被否決的替代方案、機會成本、這是誰的渴望、Pre-mortem。
Get to the root of why loadbearing instructions fail You get the Drop Table, a plain-English root cause, a rebuild, and a scoreboard you keep
Paste a decision you're stuck on — a price change, a hire, a pivot, a "should I ship this" — and get three experts argue it FOR you: the Skeptic who finds the one risk that matters, the Operator who asks what breaks in execution, the Advocate who makes the strongest case for the upside. A synthesis then names the single question that resolves it. Reasons from your input + domain knowledge only — no invented numbers, no browsing. For founders and PMs tired of arguing with themselves alone.
Run big decisions through a structured gauntlet: clarify the real question, weight your criteria, score every option, expose hidden downsides, and run a pre-mortem, without the AI making the choice for you.