15 Best ChatGPT Prompts
That Actually Work in 2026

Act as an expert research analyst. Give me a comprehensive deep-dive on the following topic: **Topic:** [[RESEARCH_TOPIC]] **Purpose:** [[WHY_YOU_NEED_THIS]] (e.g. writing an article, making a business decision, academic research) **Depth required:** [[DEPTH]] (e.g. executive summary, detailed analysis, academic level) Structure your response as follows: **1. Core Overview** (3–5 key facts every informed person should know) **2. Key Findings** (8–10 substantive insights, each 2–3 sentences) **3. Conflicting Views** (where experts disagree and why) **4. Confidence Levels** (rate each major claim: High / Medium / Low confidence, with reasoning) **5. Knowledge Gaps** (what is unknown, contested, or under-researched) **6. Key Sources to Explore** (name specific books, papers, researchers, or publications) **7. Follow-up Questions** (5 questions this research raises that I should investigate next) Be direct. Do not pad with filler. Flag speculation clearly.

I want you to argue the strongest possible case AGAINST the following position. Do not be polite or balanced — your job is to be the best possible devil's advocate. **My position / belief / plan:** [[YOUR_POSITION]] **Context:** [[RELEVANT_CONTEXT]] Rules: - Steel-man the opposition: make the opposing argument as strong as it can be - Do NOT acknowledge my position's merits during this exercise - Draw on historical examples, data, expert opinion, and logical analysis - Identify the single most fatal flaw in my position - End with: "The 3 questions you cannot afford to ignore before proceeding are..." After you've argued the opposition, give me a brief (2–3 sentence) synthesis of what this means for my decision.

Explain the following concept at two levels simultaneously, side by side: **Concept:** [[CONCEPT_OR_TOPIC]] **My current knowledge level:** [[YOUR_BACKGROUND]] (e.g. complete beginner, intermediate, know the basics but not the nuance) **Level 1 — ELI5 (Explain Like I'm 5):** Use a simple analogy a child would understand. No jargon. One clear metaphor. Maximum 100 words. **Level 2 — Expert Level:** Explain this as you would to a domain expert. Include precise terminology, edge cases, common misconceptions, and the nuance that most introductory explanations miss. Minimum 200 words. **Bridge:** In 2–3 sentences, explain what the ELI5 analogy gets wrong or oversimplifies, and why that matters for real-world application.

Help me make a high-stakes decision using a structured framework. **The decision I need to make:** [[DECISION]] **My constraints:** [[CONSTRAINTS]] (time, budget, resources, relationships) **My goals:** [[WHAT_SUCCESS_LOOKS_LIKE]] **Options I'm considering:** [[OPTIONS]] Analyze this using the following framework: **1. Clarify the Real Decision** — am I solving the right problem? **2. Options Analysis** — for each option: pros, cons, risks, estimated outcomes **3. Assumptions Check** — what am I assuming that could be wrong? **4. Reversibility** — which decisions can be undone, which cannot? **5. Second-Order Effects** — what happens 6–12 months after each choice? **6. Recommendation** — your clear recommendation with a one-paragraph justification **7. Pre-mortem** — if I take your recommended option and it fails, what is the most likely reason?

You are a brutally honest senior editor at a top-tier publication. Your job is to make this writing as strong as possible — not to be kind. **Writing type:** [[TYPE]] (e.g. blog post, essay, email, sales copy, report) **Target audience:** [[AUDIENCE]] **Goal of this piece:** [[GOAL]] (e.g. persuade, inform, convert, entertain) Here is my writing: --- [[YOUR_WRITING]] --- Provide feedback in this exact format: **Overall verdict** (1–2 sentences, honest) **Biggest structural problem** (the #1 thing that undermines the whole piece) **Line-by-line issues** (quote specific lines, explain exactly what's wrong and how to fix it) **What's actually working** (be specific, not generic) **Rewrite the weakest paragraph** to show what strong looks like **3 things to fix before publishing**

Perform a senior-level code review and refactor on the following code. **Language / Framework:** [[LANGUAGE_FRAMEWORK]] **Context:** [[WHAT_THIS_CODE_DOES]] **Main concerns:** [[YOUR_SPECIFIC_CONCERNS]] (or "full review") ``` [[YOUR_CODE]] ``` Review against these dimensions. For each issue, give severity (Critical / High / Medium / Low) and a specific fix: 1. **Correctness** — logic errors, edge cases, off-by-one errors 2. **Security** — injection risks, exposed secrets, insecure defaults 3. **Performance** — unnecessary loops, memory leaks, blocking operations 4. **Readability** — naming, function length, magic numbers, comments 5. **Error Handling** — missing try/catch, unhandled promises, silent failures Then provide the fully refactored code with inline comments explaining each significant change.

Create a detailed, personalized learning roadmap for the following skill. **Skill I want to learn:** [[SKILL]] **My current level:** [[CURRENT_LEVEL]] (complete beginner / some exposure / intermediate) **Time available per week:** [[HOURS_PER_WEEK]] hours **My goal:** [[END_GOAL]] (e.g. get a job, build a project, pass a certification, personal interest) **Learning style preference:** [[STYLE]] (videos, books, hands-on projects, structured courses, or mix) Build a 90-day roadmap with three phases: **Phase 1 — Days 1–30: Foundation** - Key concepts to master - Specific resources (named books, courses, YouTube channels, docs) - Daily/weekly practice routine - Milestone: what you should be able to do by day 30 **Phase 2 — Days 31–60: Application** - Projects to build or exercises to complete - Common pitfalls at this stage and how to avoid them - Milestone: day 60 checkpoint **Phase 3 — Days 61–90: Depth & Portfolio** - Where to go deeper - Portfolio or proof-of-skill output - Next steps after day 90 Include a "common mistakes" section for people learning this skill.

Apply first principles thinking to the following problem. Do not rely on conventional wisdom, industry norms, or "how it's always been done." **Problem or situation:** [[PROBLEM]] **Industry / domain:** [[DOMAIN]] **Conventional solution people use:** [[CONVENTIONAL_APPROACH]] Follow this process: **Step 1 — Identify assumptions** List every assumption baked into the conventional approach. Be exhaustive. **Step 2 — Challenge each assumption** For each assumption: is it actually true? Under what conditions does it break down? **Step 3 — Decompose to fundamentals** What are the irreducible physical, economic, or logical constraints? What cannot be changed? **Step 4 — Rebuild from scratch** Starting only from the fundamentals, what solution would you design if you had no legacy constraints? **Step 5 — Gap analysis** What is the gap between the first-principles solution and current reality? What would it take to close it?

Draft a professional email reply for me. **Original email I received:** --- [[ORIGINAL_EMAIL]] --- **My relationship to the sender:** [[RELATIONSHIP]] (e.g. my manager, a client, a colleague I've never met, a vendor) **Tone I want:** [[TONE]] (e.g. warm but firm, concise and professional, friendly, formal) **My main goal in this reply:** [[GOAL]] (e.g. decline politely, agree with conditions, ask for more time, push back on one point) **Key points I must include:** [[MUST_INCLUDE]] **Things I want to avoid saying:** [[AVOID]] Write the reply in full, ready to send. Then give me: - A 1-sentence note on tone choices you made - One alternative opening line if I want to sound slightly warmer or cooler

Generate exactly 20 ideas for the following challenge. Ideas must be diverse — no two should be similar in approach or angle. **Challenge / goal:** [[CHALLENGE]] **Context:** [[CONTEXT]] (industry, audience, constraints, resources available) **Ideas I've already considered (to avoid):** [[EXISTING_IDEAS]] Rules for this brainstorm: - Ideas 1–7: conventional, proven approaches (refined versions of standard ideas) - Ideas 8–14: lateral thinking — borrow approaches from unrelated industries - Ideas 15–20: provocative or unconventional — ideas that would make most people uncomfortable or skeptical For each idea: - One-line title - 2-sentence description - One reason it might fail At the end, mark your top 3 with ⭐ and explain in one sentence why each deserves serious consideration.

Process the following document and extract everything actionable. **Document type:** [[TYPE]] (e.g. meeting transcript, report, article, research paper, email thread) **My role / what I care about:** [[YOUR_ROLE]] (e.g. I'm the project lead, I'm a stakeholder, I need to brief my team) Document: --- [[DOCUMENT_TEXT]] --- Output format: **TL;DR** (3 sentences max — what this document is about and why it matters) **Key Decisions Made** (decisions that are final and not up for debate) **Action Items** | Task | Owner | Deadline | Priority | |------|-------|----------|----------| | ... | ... | ... | ... | **Open Questions** (unresolved issues that still need a decision or answer) **Watch Out For** (risks, dependencies, or concerns mentioned in the document) **What I can safely ignore** (context that doesn't require action)

Optimize my job application materials for this specific role. **Job title I'm applying for:** [[JOB_TITLE]] **Company:** [[COMPANY]] **Full job description:** --- [[JOB_DESCRIPTION]] --- **My current resume / relevant experience:** --- [[YOUR_RESUME_OR_EXPERIENCE]] --- Do the following: 1. **Keyword gap analysis** — list keywords and phrases in the JD that are missing from my resume 2. **Rewritten bullet points** — rewrite my 5 weakest bullet points to better match this role (keep them truthful) 3. **Tailored summary statement** — write a 3-sentence professional summary for this specific application 4. **Cover letter opening paragraph** — a compelling first paragraph that references this company specifically 5. **Likely screening questions** — 3 questions an ATS or recruiter will use to filter, and how I should answer each 6. **Red flags** — anything in my experience that might concern this employer, and how to address it

Act as a Socratic tutor. Your job is to help me understand the following topic through questions and guided discovery — NOT by explaining it directly. **Topic I want to learn:** [[TOPIC]] **My current understanding:** [[WHAT_YOU_THINK_YOU_KNOW]] **The specific concept I'm stuck on:** [[STICKING_POINT]] Rules you must follow: - NEVER give me the answer directly — always respond with a question - Start with what I already know and build from there - If I'm wrong, don't correct me — ask a question that reveals the contradiction - If I'm stuck, give a small hint as a question ("What would happen if...?") - Only explain something directly after I've worked it out myself - After 5–6 exchanges, ask: "What do you think you now understand that you didn't before?" Start by asking me a single question to establish my baseline understanding.

Produce a concise but substantive market analysis brief for the following. **Product / service / idea:** [[PRODUCT_OR_IDEA]] **Target market:** [[TARGET_MARKET]] **Geography:** [[GEOGRAPHY]] **Purpose of this analysis:** [[PURPOSE]] (e.g. pre-launch validation, investor pitch, internal planning) Structure: **Market Overview** Size estimate, growth trend, key drivers and headwinds (cite any known data points) **Customer Segments** 2–3 distinct buyer personas with their core jobs-to-be-done and current pain points **Competitive Landscape** 5–7 direct and indirect competitors. For each: positioning, strengths, weaknesses, approximate pricing **Market Gaps** Where are customers underserved? What pain does no current solution address well? **Go-to-Market Considerations** Best channels, common acquisition models in this space, typical CAC/LTV dynamics **Key Risks** 3 threats that could undermine success in this market **Verdict** 1 paragraph: is this an attractive market to enter, and what would it take to win?

Facilitate a structured retrospective for the following project or period. **Project / period to review:** [[PROJECT_OR_PERIOD]] **Duration:** [[TIMEFRAME]] **Team size:** [[TEAM_SIZE]] **What happened (rough summary):** [[WHAT_HAPPENED]] **Original goals:** [[GOALS]] **Actual outcome:** [[OUTCOME]] Run the retrospective through four lenses: **1. What Went Well** (be specific — name the practices, decisions, or people) **2. What Went Badly** (root causes, not symptoms — ask "why" at least twice) **3. What Was Surprising** (things you didn't anticipate, positive or negative) **4. What To Change** For each change, define: - The specific behavior/process to change - Who owns the change - How you'll know it worked (measurable signal) - When to review it **5. Key Learnings** (3 lessons that apply beyond this specific project) **6. Acknowledgments** (who deserves specific credit and for what)