The ChatGPT Instruction Library｜100+ field-tested prompts for work & creation
100+ original Traditional-Chinese-first instructions for ChatGPT / Claude / Gemini, in one place: resume reviews, mock interviews, business English email, slide outlines, meeting minutes, weekly reports, complaint replies, data readouts, social copy. Every entry ships with use-case context, variable docs, and the pitfalls we hit ourselves — paste-ready templates, not a scraped list.
- 01TUTORIAL免費TUTORIAL／ChatGPT / Claude / Gemini2026-06-12
ChatGPT Resume Health-Check Prompt: Catch AI Tells and Demand Backed-Up Numbers
This doesn't write your resume — it audits one. It flags sentences that read like AI wrote them, uses pointed follow-up questions to force you to back up vague claims with real numbers, then gives a before-and-after rewrite example plus a personal banned-words list. Use it as the last check before you submit.
BeginnerRead - 02TUTORIAL免費TUTORIAL／ChatGPT / Claude / Gemini2026-06-12
Mock Interviewer: Industry Persona, One Question at a Time, Bilingual
A turn-based mock interview: the interviewer asks scenario-based questions tailored to your industry, one at a time, follows up on your answer before giving feedback, and provides a speakable English version of the answer for each question. After eight questions, you get an overall hiring-likelihood assessment. Closer to a real interview than just reading past questions.
IntermediateRead - 03TUTORIAL免費TUTORIAL／ChatGPT / Claude / Gemini2026-06-12
Excel Formula Prompt Template: Teach the AI Your Table's Layout Before Asking for a Formula
Nine times out of ten, asking AI for a formula fails because it doesn't know what your table actually looks like. This template gives you a fill-in-the-blanks format for describing your column structure; the formula you get back comes with a step-by-step breakdown, a 5-row mini sample table, and two common gotchas flagged, with both legacy and modern function syntax provided.
BeginnerRead - 04TUTORIAL免費TUTORIAL／ChatGPT / Claude / Gemini2026-06-12
Business English Emails in Three Tones: Follow-Up, Decline, Apology
Generates the same email in three versions — formal, neutral, and friendly — each with a subject line, followed by a breakdown of "why it's worded this way." Especially useful for high-stakes emails like follow-ups, declines, apologies, and reporting up the chain — firm without burning the bridge.
BeginnerRead - 05TUTORIAL免費TUTORIAL／ChatGPT / Claude / Gemini2026-06-12
Slide Outline Generator: Page-by-Page, Ready for Gamma or Canva
Turn messy source material like meeting notes and report paragraphs into a slide outline formatted 'one point per page, separated by ---,' ready to paste straight into Gamma or Canva AI to produce a deck. Page titles are forced into statements of insight rather than labels, the conclusion goes on page two, and it comes with a 30-second opening script.
BeginnerRead - 06TUTORIAL免費TUTORIAL／ChatGPT / Claude / Gemini2026-06-12
Meeting Transcript Cleanup: Decisions, Action Items, Absentee Summary
Use this right after any speech-to-text tool: paste the transcript and get a formal record, a list of decisions, an action-item table with "Owner/Deadline/Dependencies," and a 3-line summary for people who missed the meeting. It handles transcription typos and mixed Chinese-English text, flags any deadline that wasn't actually stated as "to be confirmed," and doesn't make things up.
BeginnerRead - 07TUTORIAL免費TUTORIAL／ChatGPT / Claude / Gemini2026-06-12
Weekly Report Generator: Turn Raw Notes Into What Your Manager Actually Wants to See
Dump in a raw, whatever-you-jotted-down log of the week and get back a weekly report where results are quantified, blockers come with proposed solutions, and next week's plan is aligned to what your manager actually cares about. Scattered small tasks get merged automatically, slipping progress can't be hidden, and support requests are framed as needs — with an upward-management perspective built in.
BeginnerRead - 08TUTORIAL免費TUTORIAL／ChatGPT / Claude / Gemini2026-06-12
Boss Message Reply Generator: Handle Pressure, Pushback, and Extra Asks
For a boss's urgent demands, pushback, or last-minute extra work, this first diagnoses what they actually need (reassurance? a scapegoat? just an update?), then gives short, full, and escalation-ready reply options with risk notes for each. Built-in red-line check: it refuses goals that would require lying and offers an honest alternative instead.
IntermediateRead - 09TUTORIAL免費TUTORIAL／ChatGPT2026-06-12
ChatGPT Year-End Self-Review Generator: Notes to a KPI-Aligned Review
The worst part of writing a year-end self-review is not being able to remember what you did all year, and then having it come out reading like a diary log anyway. This prompt lets you dump in a whole batch of weekly reports, calendar events, or scattered notes, and the AI organizes them into 3-5 achievement themes, fills in quantified phrasing, and aligns everything with company goals to produce a formal self-review. It also includes a "raise negotiation" version you can use directly when discussing salary.
IntermediateRead - 10TUTORIAL免費TUTORIAL／Claude2026-06-12
Claude One-Page Proposal Prompt: Get a 3-Minute Decision
Writing a ten-page proposal your manager doesn't have time to read is the real reason most proposals go nowhere. This prompt organizes your idea into a one-page proposal—problem, solution, expected benefit, resources needed, risks—all kept to a single page, and forces the AI to push back and ask you for missing information from a decision-maker's point of view. Good for situations where you need resources, budget, or headcount from a manager or another department.
BeginnerRead - 11TUTORIAL免費TUTORIAL／ChatGPT2026-06-12
ChatGPT Customer Complaint Replies: Three Tones Plus a Compensation Call
Reply to a complaint too formally and you fan the flames; reply too submissively and the company ends up compensating for things it shouldn't. This prompt has the AI act as a senior customer service manager: it first assesses the complaint type and severity, then gives you three responses in one go — sincere apology, firm stance, and a middle-ground version — each with a "why it's written this way" note. Ready to use directly by frontline support staff, social media managers, and shop owners.
BeginnerRead - 12TUTORIAL免費TUTORIAL／ChatGPT2026-06-12
ChatGPT Cross-Team Request Emails Other Departments Will Say Yes To
Cross-team requests most often stall because "the other side has no obligation to help you." This prompt rewrites your request into a letter written from the other person's point of view: leading with the benefit or minimal burden to them, then stating exactly what you need, with a clear deadline and a fallback plan B. Outputs both a formal and a casual version in one pass, suitable for both email and Slack/Teams messages.
BeginnerRead - 13TUTORIAL免費TUTORIAL／Claude2026-06-12
Claude Data Interpretation Prompt: Numbers Into Insights Your Boss Gets
Pulling a report but not knowing how to write the 'so what' is the most common sticking point in data work. This prompt has the AI act as a senior data analyst, interpreting the data you paste in through a three-layer structure—observation, hypothesized cause, recommended action—while forcing a hard distinction between 'conclusions directly supported by the data' and 'speculation that still needs verification,' so you don't get called out in a meeting for over-interpreting.
IntermediateRead - 14TUTORIAL免費TUTORIAL／ChatGPT2026-06-12
ChatGPT Lesson Plan Generator: Topic to Script for Corporate Training
When you're assigned to run training or onboard new hires, the time-consuming part isn't knowing the content — it's arranging it into an order you can actually teach. This prompt uses the basic instructional-design framework (objectives, motivation hook, explanation, practice, assessment) to produce a complete lesson plan, including time allocation, interaction design, and points where learners tend to get stuck. Works for corporate training, new-hire onboarding, and club/community teaching.
IntermediateRead - 15TUTORIAL免費TUTORIAL／ChatGPT2026-06-12
ChatGPT Social Media Batch Prompt: 5 Distinct, On-Brand Posts in One Go
A social media manager's daily grind is finding five different ways to talk about the same product. This prompt first has you define your brand voice and no-go topics, then generates a full batch of posts from five distinct angles — pain point, scenario, data, interactive question, and behind-the-scenes — each with hashtags and a suggested posting time, plus built-in rules to strip out that "obviously AI" tone so every post doesn't read like the same canned copy.
BeginnerRead - 16TUTORIAL免費TUTORIAL／Claude2026-06-12
Claude Job Handover Generator: Extract Tacit Knowledge Into a Manual
The biggest problem with handover documents is that everything you consider obvious never gets written down. This prompt has the AI act as a handover reviewer, first using structured questions to draw out the tacit knowledge in your head (who to contact, where passwords live, what has to happen every month, where the landmines are), then organizing it into a handover manual the successor can actually follow. Useful before resigning, transferring roles, or taking extended leave.
IntermediateRead - 17TUTORIAL免費TUTORIAL／ChatGPT / Claude2026-06-04
Regex Translator: Plain Language In, Regex Plus Line-by-Line Explanation Out
No need to memorize regex syntax — describe in plain language what you want to match, and the AI gives you the regex, an explanation of what each part does, and test cases.
SAMPLE一段正則 + 逐段解釋 + 該中/不該中的測試字串。BeginnerRead - 18TUTORIAL免費TUTORIAL／ChatGPT / Claude / Gemini2026-06-04
Plain English to SQL: Describe What You Want, Get the Query
Converts plain-language requests like 'find customers who bought more than 3 times last month' into an executable SQL query plus an explanation — give it your table schema for even more accurate results.
SAMPLE一段只用你給的欄位、可執行的 SQL + 一行說明。IntermediateRead - 19TUTORIAL免費TUTORIAL／ChatGPT / Claude / Gemini2026-06-04
The Feynman Technique: Explain Any Hard Concept So a 12-Year-Old Gets It
Use the Feynman Technique to have AI teach you: first explain it clearly in plain language with an analogy, then flag the point you're most likely to misunderstand, and finish with a self-check.
SAMPLE一段大白話比喻 + 常見誤解 + 自我檢測題的教學。BeginnerRead - 20TUTORIAL免費TUTORIAL／ChatGPT / Claude / Gemini2026-06-04
Decision Matrix: Get Unstuck Between Two or Three Options
Hand the AI the options you're torn between; it lists the key considerations, gives weighted scores, points out risks you might have missed, and gives a reasoned recommendation.
SAMPLE一張評分表 + 被低估的風險 + 一個有理由的明確建議。IntermediateRead - 21TUTORIAL免費TUTORIAL／ChatGPT / Claude / Gemini2026-06-04
Cold Outreach Emails That Get Replies: A 4-Line Template That Doesn't Read Like a Sales Pitch
For emailing strangers: one line of relevant opening, one line naming their pain point, one line of concrete value, and one low-friction CTA — avoiding the 'our company was founded in...' dead-on-arrival opener.
SAMPLE一封 90 字內、開場講對方、CTA 低門檻、不像罐頭的開發信。IntermediateRead - 22TUTORIAL免費TUTORIAL／ChatGPT / Claude2026-06-04
Landing Page Copy: Using AIDA to Turn Scrollers into Buyers
Generate a full sales-page copy skeleton in one shot: an attention-grabbing headline, amplified pain points, stacked benefits, objection handling, and a strong CTA — ready to drop your product straight in.
SAMPLE一份含 headline / 痛點 / 好處 / 破疑慮 / CTA 的完整銷售頁文案。IntermediateRead - 23TUTORIAL免費TUTORIAL／ChatGPT / Claude / Gemini2026-06-04
Ad Headlines ×10: Generate 10 Different Hook Angles for One Product
Ad testing creative for Facebook / Google: generate 10 headlines from different angles (curiosity, numbers, pain points, contrast, urgency...) at once, ready for A/B testing.
SAMPLE10 個標好切角、各 12 字內的廣告標題，可直接拿去 A/B 測。BeginnerRead - 24TUTORIAL免費TUTORIAL／ChatGPT / Claude2026-06-04
YouTube Title + Thumbnail Text: Pushing CTR from 4% to Double Digits
Generates 5 title-plus-thumbnail-text combinations for a single video — the title creates curiosity, the thumbnail adds non-redundant information, and together they drive more clicks than either alone.
SAMPLE5 組標題 + 縮圖文字配對，兩者互補、合起來最想點。IntermediateRead - 25TUTORIALPROTUTORIAL／ChatGPT / Claude / Gemini2026-06-04
AEO Content: Write Passages ChatGPT / Perplexity Will Actually Cite
The new traffic battleground in 2026 is getting cited by AI. Write your topic in the format AI models love to pull from: direct answers, clear structure, and authority signals.
SAMPLE一段開頭就是可引用答案、結構化、含數據的 AEO 內容。AdvancedRead - 26TUTORIAL免費TUTORIAL／ChatGPT / Claude / Gemini2026-06-04
Sales Objection Handler: What to Say to 'Too Expensive / Let Me Think / No Time'
Gives you a response script for each of the most common customer objections — not a hard pitch, but 'acknowledge first, then reframe' — phrased naturally so it doesn't sound rehearsed.
SAMPLE針對每個異議的「認同→重框→軟收尾」自然話術。IntermediateRead - 27TUTORIAL免費TUTORIAL／ChatGPT / Claude2026-06-04
Faceless Short-Video Script: Hook People in 30 Seconds, With a Twist
A complete faceless short-video script: a 3-second hook, value in the middle, a CTA at the end, with voiceover plus shot descriptions ready to pair with AI-generated video.
SAMPLE一張 Time / 旁白 / 畫面三欄的分鏡腳本，前 3 秒有 hook。BeginnerRead - 28TUTORIAL免費TUTORIAL／ChatGPT / Claude / Gemini2026-06-04
Long-Form to X / Threads Thread: Turn One Article into a Shareable Thread
Rewrites a long article / blog post / report into a 6-9 post thread for X or Threads: the first post as the hook, one point per post, and a CTA at the end.
SAMPLE一串編號好、第一則是 hook、每則一個點、結尾有 CTA 的 thread。BeginnerRead - 29TUTORIALPROTUTORIAL／Claude / ChatGPT2026-06-04
AI Agent Task Spec: Turn Vague Requests into Executable Instructions
2026 has entered the agent era — this template turns 'I want X' into an executable spec for an agent: goal, steps, available tools, definition of done, and boundaries.
SAMPLE一段逐步執行、每步回報、卡住給 fallback、達標才停的 agent 執行。AdvancedRead - 30TUTORIAL免費TUTORIAL／Claude / ChatGPT2026-06-04
AI Code Review: Catch Bugs, Security Issues, and Readability Problems
Have AI review your code like a senior engineer: it flags concrete issues and fixes across four dimensions — correctness, security, performance, and readability — instead of vague praise.
SAMPLE一份依四面向、只列真問題、每條附行號+原因+修法的審查報告。IntermediateRead - 31TUTORIAL免費TUTORIAL／ChatGPT / Claude / Gemini2026-06-04
Personal Brand Post Batch: Generate 5 Angles on One Topic in a Single Pass
Building a personal brand doesn't mean racking your brain for a new angle every day — give it one topic, and it produces 5 social posts from different angles (story / opinion / how-to / contrast / interactive) in one go.
SAMPLE5 則不同切角、各有強開場、符合你語氣的貼文。BeginnerRead - 32TUTORIAL免費TUTORIAL／ChatGPT / Claude / Gemini2026-06-04
Meeting Notes to Action Items: Paste a Transcript, Get Who/What/When
Turns a messy meeting transcript or notes into decisions, action items (owner + deadline), and open questions — done in 30 seconds after the meeting.
SAMPLE一份含決議、行動項表（負責人/任務/期限）、待確認問題的會議摘要。BeginnerRead - 33TUTORIAL免費TUTORIAL／Gemini 2.5 Pro2026-06-04
Gemini 2.5 Pro: Drop a Screenshot, Get Your Next Steps
Uses Gemini 2.5 Pro's multimodal ability to turn any screenshot — an error message, an admin dashboard, a report, a conversation — into a concrete, actionable next-step checklist.
SAMPLE一段「看到什麼 + 最該注意的 + 3 個下一步」的可執行回覆。IntermediateRead - 34TUTORIALPROTUTORIAL／Claude / ChatGPT / Gemini2026-06-04
Context Engineering: What Separates You in 2026 Isn't the Incantation, It's the Ingredients
Feed the AI structured background, what's already been tried, constraints, and available tools/data structures before giving it the task — in 2026 the real leverage point has shifted from 'prompt tricks' to 'context.'
SAMPLE一段只根據你給的 context 作答、缺料時主動點出缺什麼的回覆。AdvancedRead - 35TUTORIALPROTUTORIAL／ChatGPT / Claude2026-06-04
Prompt Chaining: Split a Big Task into Three Steps for Better Quality
Don't feed an entire complex task into a single prompt — split it into three separate steps ('decompose → produce each part → integrate'), and the output is noticeably more stable and complete.
SAMPLE先回一個拆好的子任務清單；最後把各步整合成一份完整交付物。AdvancedRead - 36TUTORIAL免費TUTORIAL／ChatGPT / Claude / Gemini2026-06-04
Self-Critique Revision Loop: Draft, Self-Criticize, Revise — Three Rounds of Polish in One Prompt
Has the AI produce a first draft, then switch into a strict editor role to pick apart its own flaws, and finally rewrite based on that critique — effectively running three rounds of polish inside a single prompt.
SAMPLE最終改寫版 + 一行「改了什麼」說明（中間草稿與批評不顯示）。IntermediateRead - 37TUTORIAL免費TUTORIAL／ChatGPT / Claude / Gemini2026-06-04
Rubric-Driven Output: Tell AI What a Perfect Score Looks Like First
Write out the grading rubric in your head and give it to the AI first — it will use that yardstick to self-align, self-evaluate, and fix whatever falls short, getting you much closer to the quality you actually want.
SAMPLE一段已自評並把不到標處改掉的輸出。IntermediateRead - 38TUTORIAL免費TUTORIAL／ChatGPT / Claude / Gemini2026-06-04
The Anti-Goal Sandwich: One Line to Stop AI's Most Common Mistake
Explicitly naming "the most common failure mode AI makes on this type of task" in the prompt and telling it to proactively avoid that mode — 2026 field tests show roughly a 28-30% reduction in errors.
SAMPLE一段先在 <thinking> 內推 3 步、再給最終答案的回覆，且主動避開你指定的失敗模式。IntermediateRead - 39TUTORIALPROTUTORIAL／Claude / ChatGPT2026-06-04
XML Tag Structuring: 28% Fewer Parsing Errors Than Markdown
Wraps instructions, data, rules, and schemas in XML tags when feeding them to an AI — tested to produce roughly 28% fewer errors on structured-extraction tasks than using Markdown headers.
SAMPLE一段嚴格符合你給的 schema 的 JSON、不夾雜任何多餘文字。AdvancedRead - 40TUTORIAL免費TUTORIAL／ChatGPT / Claude / Gemini2026-06-04
Contrastive Example Calibration: Show It What You Want AND What You Don't — One Negative Example Beats Three Positive Ones
Demonstrate both the 'desired output' and the 'output that shouldn't appear' at the same time, so the AI calibrates between the two poles; for tasks about tone and precision, a single negative example often outperforms three positive ones.
SAMPLE一段語氣與深度貼近你正例、且避開反例毛病的輸出。IntermediateRead - 41TUTORIAL免費TUTORIAL／ChatGPT / Claude / Gemini2026-06-04
Chain-of-Thought Scaffold: A Fixed 5-Step "Think Before You Answer" Process
Forces the AI to think through 5 fixed steps before answering — restate → list assumptions → reason step by step → self-check → conclude. Field tests show roughly a 30% reduction in confidently wrong answers on complex reasoning tasks.
SAMPLE一段先在 <reasoning> 內走完 5 步、最後才給結論的回覆。IntermediateRead - 42TUTORIAL免費TUTORIAL／Claude 4.5 Sonnetr/PromptEngineering·2026-05-14
5 Claude Repeat-Task Templates That Lasted 6+ Months — and the 3 Retention Rules
After 18 months of testing 60 Claude workflows, the retention pattern: annoying but not painful, output has a destination, input under 30 seconds. The 5 templates that survived 6+ months (Friday review / meeting follow-up / client weekly report / Monday briefing / pipeline update).
SAMPLE5 個模板實際輸出：(1) Friday review 拿到誠實檢討 + 下週 top 5 + 該改的一件事 (2) Meeting follow-up 拿到 summary table 加可寄 email (3) 客戶報告 ready to paste (4) Monday bIntermediateRead - 43TUTORIAL免費TUTORIAL／ChatGPT / Claude / Geminir/ChatGPTPromptGenius·2026-05-14
5 Prompt Engineering Techniques That Actually Move the Needle (XML Beats Markdown by 28%)
Five quantifiable techniques from an AI engineer's hundreds of hours testing across GPT-4, Claude, and Gemini: CoT scaffolding, persona + goal + anti-goal, XML being 28% more accurate than markdown, negative examples outperforming positive ones, and prompt chaining beating the mega-prompt.
SAMPLE同樣 task：mega-prompt 出 cheerleading 式 mediocre 回應；加完 5 個技巧之後出 precise / qualified / no-hype / structured / 有 reasoning trail 的回應。Structured eIntermediateRead - 44TUTORIAL免費TUTORIAL／ChatGPTr/ChatGPTPromptGenius·2026-05-14
Turn Off ChatGPT's Polite Lying — 3 Anti-Sycophancy Instructions
By default, helpful = agreeable = burying the real problem in paragraph four. Three one-line instructions reverse it: problem first, force a recommendation, explicitly forbid softening. The difference between reading and fixing.
SAMPLE同一個 business idea：default 回應「This has strong merit because... one consideration might be...」；加完 3 個指令：「This has 3 problems. (1) Your unit ecBeginnerRead - 45TUTORIALPROTUTORIAL／Claude 4.5 Sonnetr/PromptEngineering (HackAPrompt 2.0 1st)·2026-05-14
A Red Teamer's View: 3 High-Capability Prompt Patterns That Work Across Models
Three patterns shared by the HackAPrompt 2.0 champion: frame the task as an audit, make the abstraction level explicit, and stage the context like production. They work on Claude, GPT, and Gemini alike.
SAMPLEAI output 從 average tutorial 級跳到 production-grade — 命名、edge case、type hints、doc strings 全部上一個檔次；audit 段會出現「真正的答案」；production-shape 輸入測過、不會「測AdvancedRead - 46TUTORIALPROTUTORIAL／Claude 4.5 Sonnetr/PromptEngineering·2026-05-14
AI Agent Harness Engineering: Separating the LLM from the Deterministic Orchestration Layer
Mitchell Hashimoto's "Agent = Model + Harness" framework. The reason 88% of enterprise AI agents never reach prod isn't the prompt — it's the missing deterministic orchestration layer. Four CS primitives you can't skip: state machine, idempotency, DAG, DLQ.
SAMPLEAI agent 在 prod 穩定運行 — task 不會被 double-execute、信用卡不會被重刷、failed task 進 DLQ 給你看不是消失、agent 100% 受 harness 控制不自己亂跑AdvancedRead - 47TUTORIAL免費TUTORIAL／ChatGPTr/ChatGPTPromptGenius·2026-05-14
COMMAND PLANNING MODE: A 9-Section Structured Planning Prompt
No conversation, no suggestions, no attempts — it just turns any situation into a structured execution plan. Fixed 9-block output: Reality / Objective / 4-phase Timeline / Tactics / Threats / Defenses / Contingency / Leverage / Confidence. Same input always yields the same output.
SAMPLE9 區塊完整計畫：[CURRENT REALITY] / [TARGET OUTCOME] / [EXECUTION TIMELINE 4 stages] / [TACTICAL ACTIONS] / [THREAT MAPPING] / [DEFENSE LOGIC] / [CIntermediateRead - 48TUTORIALPROTUTORIAL／ChatGPTr/ChatGPTPromptGenius·2026-05-14
System vuln scanner: analyze your architecture, find attack chains, estimate time-to-compromise
More than "list the CVEs." It finds how individual weaknesses chain into an attack and estimates how long an AI-augmented attacker would need to break in. A take on the OpenAI Daybreak concept, built for small teams. Feed in a stack description (Next.js + Supabase + Clerk, etc.) and get back critical paths plus a patch-priority order.
SAMPLEAudit Summary 含 critical paths 數量 + 每個發現的 severity + AI-assisted risk 評等 + attack chain 串連分析 + time-to-compromise 估計 + 分層補丁（quick fix vs strAdvancedRead - 49TUTORIAL免費TUTORIAL／ChatGPTr/ChatGPT·2026-05-14
ChatGPT Custom Instructions: describe the relationship, not your resume
Most people fill Custom Instructions with a LinkedIn resume, "I'm a software engineer, I like bullet points," and it does nothing. Swap in a description of the conversational relationship you want, and three sentences will completely change how the model behaves.
SAMPLEChatGPT 對話風格徹底轉變：你會看到 reasoning 一句先出、不確定處標 [uncertain]、它開始 disagree、開始質疑你的 framing、不再說「Great question!」開頭、不再「Would you like me to...」結尾BeginnerRead - 50TUTORIALPROTUTORIAL／Claude 4.5 Sonnetr/PromptEngineering (tobs-code — github.com/tobs-code/prompts)·2026-05-14
reClaim: Turn a Frontier Model Into a Research Agent With Confidence Scores
Scores every claim on three axes (A: sources / B: counterevidence / C: completeness), plus a 4-tier source hierarchy, adversarial self-poking, and 4 output modes.
SAMPLEStructured 信心評分 + 來源 tier 標註的答案、contradictions 不被平均掉而是攤開、有 adversarial 反證過、知道哪些可信哪些要 verifyAdvancedRead - 51TUTORIAL免費TUTORIAL／ChatGPTr/ChatGPTPromptGenius·2026-05-14
AI attack-surface audit: see yourself the way an AI-augmented attacker does
Written after Google's threat-intelligence report in May 2026. AI hacking has gone from nascent to industrial scale, with commercial AI models digging up zero-days humans had missed for decades. This prompt lays out every place you can be attacked.
SAMPLE結構化威脅評估：attack surface 表 + AI-augmented 威脅清單 + 風險矩陣 + 具體 fix 步驟（不是「用強密碼」level）+ 整體曝險評等（Low / Moderate / Significant）IntermediateRead - 52TUTORIAL免費TUTORIAL／ChatGPTr/ChatGPTPromptGenius·2026-05-14
Job-search strategy: turn ChatGPT into a 4-in-1 (auditor + analyst + advisor)
Combines a resume auditor, market analyst, system auditor, and strategist. It won't flatter you, won't over-reassure you, and won't push you toward roles that don't fit. Includes a 10-step process and ghost-job detection.
SAMPLE10-step 客製化求職策略 — 含 multi-lane resume 版本建議 / 每個職缺的 fit-risk 評分 / cover letter 量身改 / application board / ghost job flagIntermediateRead - 53TUTORIAL免費TUTORIAL／ChatGPTr/ChatGPTPromptGenius·2026-05-14
English copy editor: 4 priorities, emotional resonance, and structural flow
More than a grammar fix. Purpose-built for persuasive and narrative writing: four priorities (clarity → tone → emotional resonance → structure), an explanation for every change, plus extra suggestions.
SAMPLE清楚標出的修改後版本 + 為什麼改的 1 段說明 + 進一步精修的具體建議 — 不是「修文法」、是「為什麼這樣寫更有力」BeginnerRead - 54TUTORIALPROTUTORIAL／Claude 4.5 SonnetFEATUREDr/ClaudeAI·2026-05-14
Claude Life OS Prime: Turn the Chat Window Into a Productivity Dashboard
Not a chatbot — think Notion AI + Motion + Atomic Habits in one. Tracks habits, goals, energy, and deep work; auto-detects behavioral patterns; flags burnout risk.
SAMPLE高度個人化的 productivity 教練體驗 — 上線 onboarding 後每日 check-in 給你 score card + 1 個具體 fix、週日週報抓 cross-day pattern、月底長期趨勢 + burnout 風險評等AdvancedRead - 55TUTORIAL免費TUTORIAL／Claude 4.5 Sonnetr/ChatGPTPromptGenius·2026-05-14
A Universal Decision Prompt: Use "What Will You Regret in a Year" to Force a Verdict
Stop picking a role for every decision (financial planner / career coach / relationship coach) — one universal prompt, four steps. The regret line is the core — it forces the model from "it depends" to "actually do this."
SAMPLE4 段結構化決策分析 — 專家是誰 + 他們看什麼 + 應該怎麼做 + 大部分人哪裡錯了 + ONE-sentence 結論（給你你需要聽的話）BeginnerRead - 56TUTORIAL免費TUTORIAL／Claude 4.5 Sonnetr/PromptEngineering (Anbeeld — RESUME.md GitHub repo 作者)·2026-05-14
CV Writer Ruleset: It Interviews You STAR-Style and Bans "Dynamic Team Player"
The pain of writing a CV isn't the layout — it's listing duties instead of achievements, then having AI pile on and turn you into a "dynamic team player." This ruleset forbids AI from drafting anything until it has your evidence and positioning, and walks you through one question at a time.
SAMPLECV 帶 sharp 的成就導向 bullets（不是 responsibility list）、每個 role 至少 60% bullet 有數字或 qualitative anchor、零 weasel verb、符合目標市場 conventionIntermediateRead - 57TUTORIAL免費TUTORIAL／Claude 4.5 Sonnetr/ClaudeAI·2026-05-14
Voice Memo → Tomorrow's Priorities (With a Confirmation Round)
Before you log off, ramble into your phone mic for 2 minutes, hand it to Claude, and get back a sorted list: first thing tomorrow morning + ranked to-dos + waiting-on + can-defer. Auto-detects daily vs. weekly mode.
SAMPLE結構化日 / 週計畫、用你的語言寫的、模糊處被標 [CLARIFY]、明天第一件事一目了然BeginnerRead - 58TUTORIAL免費TUTORIAL／Claude 4.5 Sonnetr/ClaudeAI·2026-05-14
The 60-Second Pre-Send Check: Catch the Message You'll Regret Tomorrow Morning
The author wrote this after 30% of their heated replies made them wince the next day — costing a few deals and a friendship. The trick: don't let AI rewrite it, let it show you what the calm version would look like. Once you see it, you usually can't go back to the original.
SAMPLE4 段分析：意圖 vs 落地差距 / 最易誤讀的一句 / 冷靜版會長怎樣（最有用的一段）/ 送 vs 等 vs 改寫的明確建議BeginnerRead - 59TUTORIAL免費TUTORIAL／Claude 4.5 SonnetFEATUREDr/ClaudeAI·2026-05-14
5 Claude Weekly Workflows That Stuck After 6 Months
Proposal generator / meeting processor / content repurposer / Friday review / shutdown reset — out of 40 I tried, only these 5 survived, each saving 30+ minutes per run.
SAMPLE5 個 workflow 各自獨立的乾淨 output — proposal Word 文 / 會議紀要 + action items + follow-up email / 1→5 篇內容 / 週五回顧 + next-week 排序 / EOD 反思IntermediateRead - 60TUTORIALPROTUTORIAL／Claude 4.5 SonnetFEATUREDr/PromptEngineering·2026-05-14
7 Cross-Model A/B-Tested Prompt Techniques (Not Vibes)
Written after six months of A/B testing 200+ prompts on Claude, GPT-4, and Gemini with human raters. Covers persona + constraint stacking, anti-examples, role-reversal QA, cognitive scaffolds, emotional priming, uncertainty CoT, and steelman-first.
SAMPLEOutput 從「kind of works」升到「embarrassingly good」— 對複雜 review / critique / 創意生成任務最有感AdvancedRead - 61TUTORIAL免費TUTORIAL／ChatGPTFEATUREDr/ChatGPTPromptGenius·2026-05-14
5 trust-checking one-liner prompts (no role-play)
After a year of role-play system prompts and 14-step frameworks, here's the takeaway: the thing that actually moved the needle on quality was 5 one-line prompts. No roles, no markdown, no "you are an expert."
SAMPLE更可靠的 AI 答案 — 知道 model 哪裡在 hallucinate、哪裡在 self-censor、哪裡真的懂、哪裡只是看起來懂BeginnerRead - 62TUTORIALPROTUTORIAL／ChatGPTr/ChatGPTPromptGenius·2026-05-14
Grading essays as a teacher: a 3-step chain (lens → diagnose → tone)
A method dreamed up by a teacher 19 essays into a 30-essay all-nighter. Where typical AI prompts "act as an English teacher" and devolve into a wall of rubric noise, this chain solves the real grading workflow: don't comment on everything.
SAMPLE聚焦的 margin comments（每個 < 20 字、附原文引用）+ 180 字內 end note（≤ 3 個 growth points）、用你的真實語氣不像 AI、學生能真的看完並改AdvancedRead - 63TUTORIAL免費TUTORIAL／ChatGPTr/ChatGPTPromptGenius·2026-05-14
AI displacement self-assessment: break your job into tasks and score each 1-10
Written after Cloudflare cut 20%, BILL cut 30%, and Upwork cut 24% in 2026. Your job title tells you nothing; breaking the work into tasks tells you everything. Get an AI vulnerability score (1-10) and timeline for each task, plus an overall risk rating.
SAMPLE5-10 個任務分數表（1-10 vulnerability + 1-2 句理由 + timeline）+ 整體 GREEN/YELLOW/RED 風險評等 + 3 個具體 next stepIntermediateRead - 64TUTORIALPROTUTORIAL／Claude 4.5 Sonnetr/ClaudeAI·2026-05-14
Flip Your AI Dev Workflow: Generate the Prompt First, Then the Code (4 Phases)
Instead of telling AI to write code directly, have it translate a vague goal into an executable prompt structure first, then use that structure to drive code generation. Phase 1 Intent → Phase 2 Structuring → Phase 3 Tool-Level → Phase 4 Implementation.
SAMPLE4 phase 完整輸出：Phase 1 簡短意圖 / Phase 2 5 段結構（subtask + constraints + success + tool-use + checkpoints）/ Phase 3 N 個 executable prompts / PhaseAdvancedRead - 65TUTORIAL免費TUTORIAL／ChatGPTr/ChatGPTPromptGenius — ChatGPT Prompt of the Day·2026-05-14
Custom GPT governance: the 7-block Agent Identity Card
Run this before you build a custom GPT or n8n agent. It forces you to define boundaries up front and prevents agent scope creep ("just handle my work for me" = disaster).
SAMPLE7 區塊完整 markdown 表格、每區塊有具體填空、可直接複製進 custom GPT 的「Instructions」欄、agent 上線後不會 scope creepIntermediateRead - 66TUTORIAL免費TUTORIAL／ChatGPTr/ChatGPTPromptGenius — ChatGPT Prompt of the Day·2026-05-14
GPT BS detector: catch the moments AI gets it wrong just to please you
An Oxford study in Nature: the "warmer" an AI is trained to be, the more its accuracy drops, by 10-30 percentage points. This prompt reviews an AI response and flags every place warmth overrides accuracy.
SAMPLE5 區塊審查報告：warmth signals 列點 + accuracy gaps 標 ❌⚠️🚧 + 1 句 trade-off 的具體例子 + LOW/MED/HIGH 風險分級 + 不帶 padding 的乾淨改寫IntermediateRead - 67TUTORIALPROTUTORIAL／Gemini 2.5 Pro2026-05-14
Read 10 papers at once and synthesize them with Gemini
Upload 10 PDFs and Gemini automatically maps the consensus, disagreements, and gaps, then drafts a literature review.
SAMPLE5 段結構化文獻綜述、含 methodology table、引用具體 paper 名稱、誠實標 hedgeAdvancedRead - 68TUTORIALPROTUTORIAL／Gemini 2.5 Pro2026-05-14
Auto-Brief a 10-K with Gemini
Upload a public company's annual or quarterly report PDF and auto-generate a management-quality + risk read plus an investment thesis
SAMPLE4 段分析：business quality / management tone（引言）/ red flags（含 page #）/ 3 個提問AdvancedRead - 69TUTORIAL免費TUTORIAL／Gemini 2.5 Pro2026-05-14
Onboard New Hires to a Codebase with Gemini
Upload the whole repo or its key files, and Gemini gives new hires a reading order to understand the codebase within an hour
SAMPLE4 stage 讀檔計畫、每檔有「look for X」hint、加「不要看這些」防嚇跑IntermediateRead - 70TUTORIAL免費TUTORIAL／Gemini 2.5 Pro2026-05-14
Auto-generate action items from meeting recordings and video with Gemini
Upload a 60-minute meeting recording and Gemini surfaces every action item, owner, and timestamp.
SAMPLE4 段：action items table / decisions / open questions / vibe check，全有 timestamp 引言BeginnerRead - 71TUTORIAL免費TUTORIAL／Gemini 2.5 Pro2026-05-14
Reverse-Engineer a UI Bug from a Screenshot with Gemini
Paste 1-3 screenshots (browser / Figma / app) and Gemini pinpoints the visual problem + the fix
SAMPLE3 段：visual diagnosis / 3 個候選修法（含信心度）/ 還需要看什麼確認IntermediateRead - 72TUTORIALPROTUTORIAL／Gemini 2.5 Pro2026-05-14
Prioritize your product feature list with Gemini
Upload customer feedback, NPS data, or support tickets alongside your feature backlog, and Gemini runs a RICE or Kano model analysis.
SAMPLE3 段：top 5 ranked features（含 RICE math + Kano + 引言）/ NOT 做的 3 個 / 缺漏的需求AdvancedRead - 73TUTORIAL免費TUTORIAL／Gemini 2.5 Pro2026-05-14
Compare two contract PDFs and find the differences with Gemini
For legal, procurement, and HR teams: upload two PDFs and get an automatic line-by-line list of clause differences.
SAMPLE結構化 diff list、每條標 section + page + materiality + 1 句解釋、結尾 executive summaryIntermediateRead - 74TUTORIALPROTUTORIAL／Claude 4.5 Sonnet2026-05-14
Run an Architecture Review on Your Whole Codebase with Claude
Upload the repo and Claude returns an architectural map, the top 5 fixes to prioritize, and the file paths to back them up.
SAMPLE3 段式輸出：架構地圖 / Top 5 改善（含 file:line + code + 工時 + 風險）/ 不要動的 2-3 件事AdvancedRead - 75TUTORIALPROTUTORIAL／Gemini 2.5 Pro2026-04-25
Deep Document Analysis with Gemini's 1M Context
Put Gemini's 1M-token context to work — feed in an entire PDF or cross-reference multiple documents at once
SAMPLE結構化交叉分析報告、每點附來源、找跨文件的非顯性 insightAdvancedRead - 76TUTORIALPROTUTORIAL／Gemini 2.5 Pro2026-04-25
Deep multimodal image analysis with Gemini
Upload an image for structured analysis: design reviews, UX audits, and competitive screenshot comparisons.
SAMPLE結構化視覺批評、引用具體元素、不模糊建議IntermediateRead - 77TUTORIALPROTUTORIAL／Gemini 2.5 Pro2026-04-25
Whole-Codebase Overview with Gemini
Drop the entire repo into Gemini → get an architectural overview + improvement suggestions
SAMPLEArchitectural map + 5 件優先改善 + 引用檔案路徑、無 hallucinateAdvancedRead - 78TUTORIALPROTUTORIAL／ChatGPT2026-04-24
2,000-Word SEO Blog Post
An SEO-optimized 2,000-word blog post — target keyword research, H2/H3 structure, internal links, and meta description all generated in one pass.
SAMPLE產出 2000 字完整 SEO blog post、title / meta / H1-H3 / FAQ 全部到位、keyword 自然分佈、internal link placeholder 標清、可直接丟進 CMS 發表。IntermediateRead - 79TUTORIAL免費TUTORIAL／ChatGPT2026-04-24
Social Post Formulas for IG/LinkedIn
Give ChatGPT your industry and your point of view, and it produces 5 social posts using 5 viral formulas — Contrarian, List, Story, Question, and Data.
SAMPLE產出 5 篇社群貼文、每篇不同公式、各帶 hashtag + CTA 建議、字元數符合平台。一週內容就搞定。BeginnerRead - 80TUTORIAL免費TUTORIAL／ChatGPT2026-04-24
Viral Headline Generator: 10 Directions
Give ChatGPT your content topic, and it produces 10 directions using 10 viral headline formulas. A/B test them or just pick the best fit.
SAMPLE產出 10 個不同公式的標題、各標 CTR 估計與建議使用場合。可直接拿 2-3 個最好的上平台 A/B。BeginnerRead - 81TUTORIALPROTUTORIAL／ChatGPTFEATURED2026-04-24
7-Email Nurture Sequence
A 7-email automated sequence for new subscribers — welcome, education, trust-building, community, a small offer, the main offer, and a win-back email. Copy straight into ConvertKit or Mailchimp.
SAMPLE產出 7 封完整 email、每封有 subject / preheader / body / CTA、語氣一致、結構漸進推到第 6-7 封的 offer。直接貼 ConvertKit / Mailchimp 排程、上線。AdvancedRead - 82TUTORIAL免費TUTORIAL／ChatGPT2026-04-24
Interview Simulator: Turn-Based Q&A
An interactive interview simulator — ChatGPT plays the interviewer, asking questions and follow-ups based on the job description and your role, then giving feedback. Ten times more effective than practicing alone talking to yourself.
SAMPLE啟動一場 8 問互動面試、每題有 follow-up 追問、最後給完整診斷 + 改進建議。練 30 分鐘等於跟真 recruiter 對話一輪。IntermediateRead - 83TUTORIAL免費TUTORIAL／ClaudeFEATURED2026-04-24
Resume Tailoring for a Specific Job Description
Feed Claude your original resume plus the target job posting, and it rewrites every section of your resume in that company's language — keywords, metrics, and industry terms all aligned — to boost your ATS pass rate.
SAMPLE產出重寫後的履歷 Markdown — 每段工作經歷都 rephrase 過對齊 JD 用語、關鍵字密度提升但不生硬、metric 換成對那家公司有意義的指標、ATS 友善格式（無表格、無圖）。直接另存 PDF 投履歷。BeginnerRead - 84TUTORIALPROTUTORIAL／ClaudeFEATURED2026-04-24
Interview Transcript to Story Synthesis
Turns a 3-hour interview recording into a 2,000-word brand story, profile piece, or documentary script. Claude selects the key passages, preserves the original voice, and stitches fragmented moments into a narrative with a real arc.
SAMPLE產出一篇有新聞文學質感的長文、開頭有具體場景不是摘要、中段保留 3 個以上受訪者原語引用、中間點出一個 tension、結尾 callback 開頭場景、讀起來像 New Yorker profile 而非 AI 合成文。AdvancedRead - 85TUTORIALPROTUTORIAL／Claude2026-04-24
1-Page Decision Memo for the CEO
A one-page decision memo for a CEO, board, or VC. Compresses 30 pages of analysis into a single clear document readable in 5 minutes before a meeting, in the style of Amazon's PRFAQ or McKinsey's SCQA.
SAMPLE產出 1 頁 Markdown / Word 備忘錄、6 個段落順序結構化、BLUF 第一句就看到結論、表格比較 3 個選項、每個 claim 有 [Source] 標註、結尾明確問「你要對方做什麼」。11pt 印出來剛好一張 A4。AdvancedRead - 86TUTORIALPROTUTORIAL／Claude2026-04-24
One-Page Product PRD Draft
A first-draft Product Requirements Document: a 3-sentence problem statement → 5 user stories → an API contract → success metrics. A spec engineers can start building from immediately.
SAMPLE產出 1 頁 Markdown PRD、結構化 6 段、user stories 按 P0/P1/P2 排、API contract 工程師能看懂、success metrics 具體可測、open questions 列明要決定的事。可直接開會討論、不用再大改。IntermediateRead - 87TUTORIAL免費TUTORIAL／Claude2026-04-24
Batch E-Commerce Product Copy for 30 SKUs
Give Claude basic data for 30 SKUs (name, category, selling points, materials) and get back 30 complete sets of 'title + store summary + 3 bullet features + description paragraph' in one pass—a 10x speedup for small brands listing products.
SAMPLE產出 30 組完整商品文案、格式統一可批次上架、標題優化目標平台搜尋、bullet 強調好處不是規格、段落有掃讀優化、整體語氣跟品牌故事一致。Copy-paste 到 Shopify / 蝦皮後台即可上架。BeginnerRead - 88TUTORIAL免費TUTORIAL／Claude2026-04-24
STAR Interview Answer Bank Generator
Give Claude the target job description plus 5 real work incidents of yours, and it produces 5 STAR-format interview answers you can rehearse the night before the interview.
SAMPLE產出 5 組 150 字 STAR 答案、每組都連結到 JD 的特定需求、Action 段全用「我」不用「我們」、Result 有具體數字、Tie-back 收尾把經歷包裝給那家公司想聽的樣子。背熟後面試回答流暢。IntermediateRead - 89TUTORIALPROTUTORIAL／Claude2026-04-24
Manager-Level Quarterly Self-Review Document
Give Claude a summary of your emails, meeting notes, and Slack messages from the past 90 days, and it produces a 'quarterly self-review'—achievements, failures, lessons learned, and next quarter's goals. Done 30 minutes before your performance review.
SAMPLE產出 600-800 字季度檢討、六段結構清楚、Accomplishments 都有具體數字、Struggles 段誠實但不自貶、Next Quarter 承諾可量測。可直接貼進公司 review 系統或 1:1 文件。IntermediateRead - 90TUTORIAL免費TUTORIAL／Claude2026-04-24
Professional Customer Complaint Reply
A complaint email lands and you need to reply within 5 minutes. Claude drafts the reply using a four-step framework—empathize → take ownership → offer concrete compensation → prevent recurrence—so the customer feels heard instead of escalating to a 1-star Google review.
SAMPLE產出 150-250 字回信、4 段式結構、引用客訴原話讓對方感覺被讀過、具體補償有明確 deadline、預防措施 1 句話說清楚。客戶收到會覺得「這家有在乎」— 95% 會撤回負評。BeginnerRead - 91TUTORIALPROTUTORIAL／Claude2026-04-24
Podcast Interview Outline Designer
Give Claude the guest's bio, the show's theme, and the angle you want to dig into, and it produces a complete 60-minute interview outline—15 main questions, backup follow-ups, and an opening/closing design.
SAMPLE產出完整 60 分鐘提綱、15 個主問題 + 追問、每題附「為什麼問」的心理學註解、有明確 Act 結構、結尾有 signature sign-off question。你照著讀就能錄出一集有深度的訪談。IntermediateRead - 92TUTORIAL免費TUTORIAL／ClaudeFEATURED2026-04-24
Brand About Page Story Copy
Killer copy for a brand's About / About Us page. Compresses 'who I am, why I started this business, what I believe, who this is for' into about 400 words that read like a real person, not a press release.
SAMPLE產出 400 字 About 頁文案、第一句話是具體場景不是口號、通篇第一人稱、結尾有明確的 invitation 給特定讀者。貼到網站 About 頁後預期提升信任度 — 配合 founder photo 轉換率可拉 10-15%。BeginnerRead - 93TUTORIAL免費TUTORIAL／ChatGPTu/Pixel_Pirate_Moren·2026-04-24
One Prompt That Mapped My Entire Personality (39 Traits, 6 Layers)
No quiz. No questions. I fed my old chats (things I'd written across years without ever trying to be self-aware) to ChatGPT and ran a prompt that went deeper than any personality test I've tried. 39 traits across 6 lay
SAMPLE39 個 personality traits 跨 6 個 layer（cognitive / emotional / social / motivational / behavioral / values）、附 evidence quotes 來自你過去 chatsBeginnerRead - 94TUTORIAL免費TUTORIAL／ChatGPTu/Tall_Ad4729·2026-04-24
ChatGPT Prompt of the Day: Build AI Agents That Actually Work
I've wasted more hours than I want to admit debugging AI agents that kept going off-script. Switched LLMs, swapped tools, rewrote the logic — turned out the problem was the system prompt the whole time. Too vague, too cr
SAMPLEClean agent system prompt — 含 role / context / tools / boundaries / failure modes / output format、agent 不再跑偏BeginnerRead - 95TUTORIAL免費TUTORIAL／ChatGPTu/Tall_Ad4729·2026-04-24
ChatGPT Prompt of the Day: The Ghost Job Detector
I applied to a role for three weeks. Recruiter calls, a technical screen, all of it. Then it vanished. The company kept reposting it every 30 days but nobody responded to my final follow-up. Took me an embarrassingly lon
SAMPLEGhost job 風險評等（HIGH/MEDIUM/LOW）+ 證據 checklist（reposted N 次 / 公司頁面有無 / Glassdoor 訊號 / recruiter 反應）BeginnerRead - 96
PROTUTORIAL／Claude Opus 4.72026-04-22Claude Resume Tailor: One Resume, Ten Company Applications
You have one generic resume and want to apply to 10 companies at once. Claude reorders and re-emphasizes it for each job description without changing any facts.
SAMPLEClaude 回你三區：左邊是 3 句的 summary，第一句就寫“7 年 B2B SaaS 產品經理，主導三個 0→1 商用動額設並透過活化 funnel 與定全點謟聸全傳商單收入 『天成長 8.3x 』”等要點，交廣告辦公體比又又小。工作經歷 bullet 一開始直接讓 JIntermediateRead - 97
免費TUTORIAL／ChatGPT2026-04-22ChatGPT LinkedIn Post Rewriter: Earn Reach Through Genuine Effort
You wrote up a LinkedIn reflection, but it came out reading like stale, generic corporate speak. This prompt has ChatGPT rewrite it into a version that's "long, but keeps your thumb from stopping the scroll."
SAMPLEChatGPT 給你 3 個進階版本：故事型開頭是「三年前我接了一個 8 萬元的提案，結果三個月內亟了 12 萬」；教訓型是「我以為「「需求」」是客戶說什麼我就記什麼。全錯」；問題型是「你也看過實際上是預算抱怨，但其實是兴奥限定的客戶嗎？」。三個段落都跟你原意一致、不多不少，但看BeginnerRead - 98
PROTUTORIAL／Claude Opus 4.72026-04-22Claude Code Review Assistant: Quality Check Before Merging a PR
Not a casual 'take a quick look at this code' request. One command yields a full review of the PR across four dimensions: scope/impact, security, performance, and readability.
SAMPLEClaude 給你一頁 review、規規矩矩加 4 個面向。正確性那邊他接出 2 個「必修」：你在跨 timezone 的邏輯裡用 new Date() 預設 locale，生產上會以 UTC 足誤；另一個是 rate limit 沒考慮並能取得請求，有 race conditIntermediateRead - 99
免費TUTORIAL／Claude Opus 4.72026-04-22Claude Customer Email Draft: Handle Tough Emails in 10 Minutes
For dealing with an angry customer, a complaint, or a refund request. Claude drafts a reply that's neither servile nor combative—empathetic, but not one that hands out refunds carelessly.
SAMPLE你會得到兩個版本：版本 A（較溫暖）開頭「謝謝您寫信來讓我們知道，由貪看得出來這次的經驗讓您很失望」→承認包裹碰撞是我們這邊誤標，已跟廚房舉報→「我們全額退款、並在週五前重寄一份新的」→結尾「您需要做的是等我們週五前的關櫢簡訊，裡面有退款收據與新包裹版本」。版本 B（冷靜簡潔）直BeginnerRead - 100
免費TUTORIAL／ChatGPT / Claude2026-04-20Weekly Review Journaling Prompt: Think Through Next Week in 10 Minutes
Built for freelancers, founders, and PMs. Uses a one-question-at-a-time dialogue instead of a fill-in-the-blank template to draw out genuine reflection.
SAMPLE一次丈問後 AI 回應範本：「這週你最挑戰的一次對話是哪個？花 30 秒重新感受那個場景，現在幫我數出身體哪三個部位最反應？…你提到肩重、拳頭緊、呼吸變涺——這是身體在替你記壽的情緒。在理智抵達前，身體已知道這件事跲踐到了你的哪個內在後陣？」——絕對不是「你這週覺得如何？」這種滑BeginnerRead - 101
免費TUTORIAL／Claude / ChatGPT2026-04-20Prompt Optimizer: Turn a Bad Prompt into a Good One
Give the AI your rough prompt, and it turns it into a structured, advanced prompt — and explains why.
SAMPLE輸入「打一張日本風的圖，要柔一點」，優化器輸出：「宅邸的恩居窗邊場景，晴朗下午四點半的藝術門柔光從柵水邊射入，木質地板有日光落下的不規則光斑，一塊班棉織肌理敷壊擺放，枉樹盆栽與皮件歷史小說在三寸處補足特寫距離—— 取景 85mm、f/2.0、采光自然光 —— Kodak PortBeginnerRead - 102
PROTUTORIAL／Claude Opus 4.72026-04-19Customer Interview Synthesizer: 10 Interviews Into 3 Insights
For product managers and founders. Strips the noise out of multiple interviews and extracts the shared patterns and pain points.
SAMPLE10 份訪談輸入 → 三個核心洞察：(1) 貼紙的 7/10 反映設定流程不是抽象難懂，而是「不相信會有人改」的學習出來的免疫；(2) 高付費錯誤假設—使用者不是「付不起」而是「不愿意為這個結果付」；(3) Aha moment 在第三次使用時出現—實驗設計 onboardingAdvancedRead - 103
免費TUTORIAL／Claude Opus 4.72026-04-18Meeting Recording to Three-Tier Structured Notes
Condenses a long meeting transcript into three tiers: decisions, action items, and open questions.
SAMPLE三層結構產出：【決議】Q2 下送金流切流通美元訂單到 Stripe、廣告預算以 CAC 為上限；【行動項】 @Guan 4/28 前出一份国外付款流程設計、@Mia 4/30 前完成 Stripe onboarding、司法介入 5/5；【懸而未決】是否產品語言要開英文版，待下週BeginnerRead - 104
PROTUTORIAL／Claude Opus 4.72026-04-16Claude Storyboard Auto-Generator
Give it a topic, length, and platform, and it instantly produces a storyboard plus prompts ready to feed straight into Sora or Runway.
SAMPLE「台中新婚夫妻節約裝潢」相關的 60 秒分鏡表：鏡頭 1 — 0:00-0:05 客廳闀械高角俯拍、字幕「節約 NT$40萬完成。」；鏡頭 2 — 0:05-0:12 夫妻兩人正面坐在沙發介紹；鏡頭 3 — 0:12-0:25 單項裝潢前後對比…...對應的 Sora prompIntermediateRead - 105
免費TUTORIAL／ChatGPT / Claude2026-04-01ChatGPT: The Golden 3-Second Hook for Short-Form Video
A hook formula that stops viewers from scrolling past in the first 3 seconds. One prompt produces 10 versions.
SAMPLE一次產出 10 版開場，模式更多元：「你知道 90% 的人第一個月就偵授了嗎？」、「我已經 3 天沒睡了——只為了發現一件事。」、「如果你的 IG 沒有這一個欄位……」等。每一個版本都配了 2-3 秒的帶入视覺提示，可直接 copy 貼到脚本檔。BeginnerRead - 106
免費TUTORIAL／ChatGPT / Claude2026-03-28AI Teaching Tool: Translate Jargon into Explanations Your Grandma Can Follow
Turn any complex concept into explanations at three difficulty levels with one click. Great for writing educational content or teaching AI to older relatives.
SAMPLE對「量子纏結」產出三層版本：(1) 八歲版〔兩只讓座在個別盒子裡的小牽手娃娃，一個搖頭另一個也會跟著搖頭〕、(2) 非資工版〔兩個粒子一旦操作成特殊狀態後，不管分開多遠，對其中一個的測量會立刻決定另一個的狀態〕、(3) 專業版〔Bell 不等式驗證的非局域關聯...〕。三層都保留BeginnerRead