Claude in the Real World

I'm going to share a [essay/article/email/report]. Before I share it, I want to set your role clearly: you are a direct, honest editor. Your job is not to encourage me — it's to make the writing better. When you read it, tell me: 1. What is and isn't working in the first paragraph (this is where most readers decide whether to continue) 2. Where the argument or narrative loses momentum 3. Any claims that need better support 4. One thing I should cut entirely 5. One thing that deserves more development Then give me an overall assessment of whether this piece is ready to share, needs significant revision, or needs rethinking from the ground up. Here's the piece: [paste your writing]

I'm writing a college application essay for [school]. The prompt is: "[prompt]" Before you help me write anything, ask me three questions that will help you understand what's most authentic and compelling about my experience with this topic. After I answer, reflect back to me what you think the core of my story is — the specific moment, insight, or transformation that should anchor the essay. Only after we've agreed on the core should we start drafting.

Edit the following text for concision. Your goal is to reduce it by approximately 30% without losing any substantive content, argument, or nuance. Do not make it sound like someone else wrote it — preserve my voice. Show me the edited version first, then briefly explain the most significant cuts you made. [paste text]

I need to understand [topic] deeply. Structure your response as follows: 1. Core concept (2-3 sentences that capture the essence) 2. Why this matters right now (the current relevance) 3. The key debates or disagreements within this space 4. What most people get wrong about it 5. The 3-5 most important things to know if you're going to engage seriously with this topic 6. What I should read or explore next Be direct. If something is genuinely contested, say so rather than picking a side. If there are things you're uncertain about, flag them.

I hold the following belief: [state your belief or position clearly] I want you to do two things: First, steelman this position — make the strongest possible case for it, including arguments I may not have considered. Second, make the strongest possible case against it — not a weak strawman, but the most rigorous opposition. Then tell me: based on this analysis, what would it take to change this belief, and what evidence or arguments should I be looking for?

I need to [describe what the code should do]. Context: - Language: [Python/JavaScript/etc.] - Environment: [where this will run] - Existing code it needs to work with: [describe or paste relevant code] - Constraints: [any limitations — libraries available, performance requirements, etc.] Write the code, then: 1. Explain what each significant section does 2. Identify the most likely failure points and how to handle them 3. Suggest one test case I should run before using this in production

Review this code as a senior engineer doing a security-focused code review. I need you to be thorough and specific: 1. Security vulnerabilities — any inputs that aren't validated, data that isn't sanitized, or attack vectors 2. Logic errors — cases where the code doesn't do what it appears to intend 3. Edge cases — inputs or conditions that would cause unexpected behavior 4. Performance issues — anything that would be unnecessarily slow or resource-intensive at scale 5. Readability and maintainability — anything that makes this harder to maintain than it needs to be For each issue, tell me the severity (critical/high/medium/low) and suggest the specific fix. [paste code]

I'm deciding between these options: [describe options clearly] Context: [relevant background, constraints, goals] Help me think through this systematically: 1. What am I actually optimizing for? (Reflect back what you think my real priority is based on the context) 2. What information am I missing that would significantly change this decision? 3. What does each option look like at best case, realistic case, and worst case? 4. What would I regret not having considered? 5. Your recommendation, and the one thing that could make that recommendation wrong Don't tell me what I want to hear. Tell me what I need to know.

I need to have a difficult conversation with [role/relationship] about [topic]. Here's the situation: [describe] Help me prepare by: 1. Identifying the core thing I need to communicate (distilled to one sentence) 2. Anticipating the 3 most likely objections or reactions, and how I should respond to each 3. Suggesting how to open the conversation in a way that's direct but not defensive 4. Identifying anything in my framing that might make this conversation harder than it needs to be Then give me a draft opening statement I could use or adapt.

I want to achieve: [goal] Timeframe: [when] Resources I have: [time, money, skills, access] Constraints: [what's limiting me] Don't give me a generic framework. Give me a specific, ordered action plan with: - The first 3 things I should do this week - The biggest obstacle I'll hit and how to get past it - One thing I'm probably underestimating about this - A check-in question I should ask myself at the 30-day mark If my goal is unrealistic given my constraints, tell me that directly and suggest what is realistic.