50 Copy-and-Paste Claude Prompts for Everyday Use

A practical prompt library is the fastest way to get leverage out of AI. This guide provides 50 copy-and-paste Claude prompts covering writing, coding, research, business strategy, and everyday life, along with reusable frameworks to improve your AI automation workflows.

TL;DR

• Copy, paste, and adapt 50 ready-to-use Claude prompts across five core categories: writing, coding, research, business, and personal.
• Use the universal "Quick Prompt Upgrade" block to force Claude to ask clarifying questions and separate facts from assumptions.
• Apply reusable prompt frameworks like the Context Stack and Critic Pass to eliminate generic AI output.
• Always feed Claude real context—documents, transcripts, and constraints—rather than open-ended requests.

How to use this prompt guide

Each Claude prompt in this library is designed to be used immediately to accelerate your workflows.

1. Copy the full prompt text.
2. Replace every item inside [BRACKETS].
3. Add relevant context, examples, constraints, and source material.
4. Review the output before using it.
5. Ask Claude to refine, challenge, or reformat the result when needed.

Better context usually creates better output. Include the goal, audience, constraints, examples, and desired format whenever they matter.

The Quick Prompt Upgrade

Add this text block after any prompt when you want a stronger, more structured result from Claude:

text Before answering:

1. Identify any missing context that would materially improve the result.
2. Ask only the necessary questions.
3. Separate facts, assumptions, and recommendations.
4. Use clear structure and specific examples.
5. Do not invent claims, numbers, sources, or details.
6. End with the most useful next action.

Writing & Content Creation Prompts

Ruthless Editor

text Cut this by 30 percent without losing any meaning.

Then:

• list every sentence or section you removed
• explain why it was safe to remove
• flag any claim that became less precise
• preserve the original voice

Text: [PASTE]

Voice Match

text Study the rhythm, sentence length, vocabulary, structure, and quirks in this sample I wrote.

Then rewrite [DRAFT] so it reads as if I wrote it.

Rules:

• preserve the meaning
• do not imitate errors or awkward phrasing
• avoid generic AI language
• explain the 5 strongest voice patterns you followed

Writing sample: [SAMPLE]

Draft: [DRAFT]

Hook Stack

text Write 10 opening lines for [TOPIC].

Use a different hook mechanism for each:

1. curiosity gap
2. contrarian claim
3. specific observation
4. direct callout
5. mistake
6. outcome
7. tension
8. identity
9. surprising comparison
10. practical challenge

Label each hook and keep every line under 18 words.

Reverse Outline

text Reverse-outline this draft into the argument it is actually making.

For each paragraph, provide:

• the paragraph's purpose
• the claim it makes
• the evidence it uses
• whether it advances the argument
• what should be cut, moved, or clarified

Then show the revised argument structure as a clean outline.

Draft: [PASTE]

Reader Roast

text Read this as a skeptical stranger with limited patience.

Tell me:

• the exact line where you would stop reading
• why attention drops there
• which claims feel vague or unearned
• what feels repetitive
• the smallest edits that would keep you reading

Text: [PASTE]

Clarity Pass

text Rewrite this so an intelligent 12-year-old understands it on the first read.

Keep every important fact.

Also:

• remove unnecessary jargon
• define technical terms that must remain
• shorten long sentences
• flag anything that is still ambiguous

Text: [PASTE]

Analogy Engine

text Explain [CONCEPT] using 3 fresh analogies from different domains.

Use:

• one everyday-life analogy
• one business or systems analogy
• one technical or scientific analogy

For each analogy:

• explain where it maps well
• explain where it breaks down
• tell me which one will work best for [AUDIENCE]

Headline A/B Testing

text Write 8 headlines for [TOPIC].

For each headline, score:

• clarity from 1 to 10
• curiosity from 1 to 10
• specificity from 1 to 10
• credibility from 1 to 10

Present the results in a table.

Then:

• select the best headline
• explain why it wins
• write 3 refined variants of the winner

Tone Shifter

text Rewrite this message in 4 tones:

1. warm
2. blunt
3. playful
4. executive

Present them in a table with:

• tone
• rewritten message
• when to use it
• possible downside

Message: [PASTE]

Fact Guardrail

text Rewrite [DRAFT] for clarity and flow.

Rules:

• tag every factual claim that needs verification with [CITE]
• tag every uncertain inference with [ASSUMPTION]
• tag every opinion with [OPINION]
• do not invent a number, quote, result, feature, source, or example
• preserve claims that are already supported by the provided material

Draft: [DRAFT]

Coding & Debugging Prompts

Rubber-Duck Debug

text Walk through this code line by line.

For each relevant line:

• explain what it should do
• explain what it actually does
• track important variables and state changes
• stop at the first meaningful mismatch
• identify the smallest valid fix

Code: [CODE]

Root Cause First

text Analyze this error and code.

Return:

1. root cause in one sentence
2. minimal fix
3. corrected code
4. why the fix works
5. how to prevent this class of bug
6. one test that would have caught it

Error: [ERROR]

Code: [CODE]

Screenshot Debug

text Review the attached screenshot of the broken interface and the console output.

Identify:

• what is visibly failing
• the most likely technical cause
• the evidence supporting that diagnosis
• the smallest fix
• what additional information would confirm it

Screenshot: [SCREENSHOT]

Console output: [PASTE]

PR Reviewer

text Review this diff like a senior engineer deciding whether to approve it.

Flag:

• bugs
• security risks
• broken edge cases
• performance issues
• maintainability problems
• missing tests

Rank findings by:

• blocker
• high
• medium
• low

For each finding, include a specific fix.

Diff: [DIFF]

Test-First Development

text Before writing implementation code, generate a complete test plan for [FUNCTION].

Include:

• normal cases
• boundary cases
• invalid inputs
• empty inputs
• state-related cases
• concurrency or timing cases where relevant
• security cases where relevant

Then implement the function so it passes the tests.

Requirements: [REQUIREMENTS]

Safe Refactor

text Refactor this code for readability and maintainability with zero intended behavior change.

Return:

• the refactored code
• a before-and-after diff
• an explanation of every meaningful change
• any behavior that could accidentally change
• tests required before merging

Code: [CODE]

Codebase Map

text Map this unfamiliar code.

Explain:

• its purpose
• main functions and classes
• input and output flow
• dependencies
• state changes
• error handling
• where [FEATURE] is implemented
• where you would safely begin modifying it

Code: [CODE]

Regex Builder

text Build a regular expression that matches [PATTERN].

Provide:

• the regex
• the language or engine it targets
• 3 valid examples it accepts
• 3 invalid examples it rejects
• a token-by-token explanation
• known limitations
• a safer non-regex alternative if one is more appropriate

Performance Triage

text Review this code and identify the 3 most likely performance bottlenecks.

For each:

• explain the bottleneck
• estimate likely impact
• estimate implementation effort
• give the specific fix
• explain how to measure whether the fix worked

Rank them by impact versus effort.

Code: [CODE]

Commit Surgeon

text Turn this messy diff into a sequence of clean atomic commits.

For each commit:

• list the files or changes included
• explain why they belong together
• write a conventional commit message
• note dependencies on earlier commits

Then provide the recommended commit order.

Diff: [DIFF]

Research & Analysis Prompts

Steelman First

text Steelman the strongest version of [POSITION].

Then:

• present the strongest counterargument
• list the best evidence for each side
• identify weak assumptions on both sides
• explain what evidence would change the conclusion
• give a calibrated verdict based on the available evidence

Assumption Hunt

text List every meaningful assumption in this argument.

For each assumption:

• state it clearly
• classify it as explicit or hidden
• explain why the argument depends on it
• rate how fragile it is
• explain what happens if it is false

Mark the assumptions that would collapse the conclusion.

Argument: [PASTE]

Document Interrogator

text Answer only from the document below.

Analyze:

• what it explicitly claims
• what evidence it provides
• what it implies but does not prove
• what it avoids
• what is ambiguous
• what a strong critic would challenge first

Do not add outside facts.

Document: [PASTE]

Comparison Grid

text Compare [OPTIONS] for [GOAL].

First identify the criteria that materially affect the decision.

Then create a table with:

• criterion
• importance
• each option's performance
• trade-offs
• uncertainty

Finish with:

• best option by use case
• one-line verdict
• what additional information could change the choice

Number Check

text Sanity-check every number and quantitative claim in this material.

For each one:

• reproduce the claim
• check the arithmetic
• identify missing denominators or timeframes
• flag unsupported precision
• flag contradictions
• show the reasoning
• mark anything needing a source with [CITE]

Material: [PASTE]

Second-Order Analysis

text Analyze the second- and third-order consequences of [DECISION].

Cover:

• immediate effects
• downstream operational effects
• incentive changes
• behavioral adaptations
• unintended consequences
• reversibility
• who benefits
• who bears the cost
• what people are likely to miss

Source Synthesis

text Synthesize these sources.

Create sections for:

• points of agreement
• direct conflicts
• differences in definitions or assumptions
• strongest evidence
• weakest evidence
• questions none of the sources answer
• practical conclusion

Do not force agreement where the sources conflict.

Sources: [SOURCES]

Three Levels of Explanation

text Explain [TOPIC] at 3 levels:

1. curious beginner
2. working professional
3. informed skeptic

At each level include:

• explanation
• example
• important nuance
• common misunderstanding
• one question the reader should ask next

Change My Mind

text I currently believe:

[BELIEF]

Help me examine it.

Identify:

• what evidence would strengthen it
• what evidence would weaken it
• what evidence would falsify it
• where I should realistically look
• what cognitive biases may be affecting me
• what a reasonable temporary position would be

Extract Structure

text Extract the structure from this messy text.

Create a clean table containing:

• people
• organizations
• dates
• deadlines
• decisions
• commitments
• action items
• owners
• dependencies
• unresolved questions

Then provide a short summary of what matters most.

Text: [PASTE]

Business & Strategy Prompts

Project Pre-Mortem

text Assume it is 12 months from now and [PROJECT] has failed.

Write an honest pre-mortem covering:

• the most likely reasons
• early warning signs
• assumptions that failed
• execution mistakes
• market or customer issues
• team or operational issues
• what could have been changed today

Finish with a prioritized prevention plan.

Pricing Angles

text Give me 5 ways to price [PRODUCT].

For each option include:

• pricing model
• example structure
• customer psychology
• advantages
• disadvantages
• operational complexity
• who it fits
• when it fails

Then recommend the best option for a [CUSTOMER] who prioritizes [PRIORITY].

Cold Email Fix

text Rewrite this cold email to improve the chance of a relevant reply.

Rules:

• remove filler
• lead with the recipient's likely problem
• make the relevance specific
• use one clear ask
• avoid fake personalization
• keep it concise

Show:

1. original weaknesses
2. revised email
3. why each change helps

Email: [EMAIL]

Positioning Sharpener

text Turn this vague pitch into a clear positioning statement.

Use this structure:

For [WHO] who [PAIN], [PRODUCT] is the [CATEGORY] that [EDGE].

Create 5 options.

For each option, explain:

• target customer
• core pain
• category
• differentiated edge
• possible credibility gap

Original pitch: [PASTE]

Notes to Actions

text Turn these meeting notes into 3 structured sections:

1. decisions made
2. action items with owner and deadline
3. open questions

Also identify:

• dependencies
• risks
• unclear ownership
• contradictions
• follow-ups that need scheduling

Notes: [NOTES]

Objection Bank

text List the top 8 objections a [CUSTOMER] may raise against [OFFER].

For each objection provide:

• what they are really worried about
• a concise response
• evidence needed
• a question to ask before responding
• when the objection is valid

Avoid manipulative or dismissive rebuttals.

Competitor Teardown

text Analyze this competitor material.

Extract:

• target customer
• category
• core promise
• positioning
• proof
• offer structure
• pricing signals
• strengths
• weak spots
• underserved needs
• a credible gap where I could differentiate

Material: [PASTE]

KPI Tree

text Break [GOAL] into a KPI tree.

Show:

• primary outcome metric
• leading indicators
• lagging indicators
• controllable inputs
• dependencies
• potential gaming risks
• measurement frequency

Then identify the 2 levers I can most realistically move this quarter.

Decision Memo

text Turn this material into a one-page decision memo.

Use this structure:

• situation
• decision required
• options
• evaluation criteria
• recommendation
• reasons
• risks
• mitigation
• unresolved questions
• next action

Material: [PASTE]

Negotiation Prep

text Prepare me to negotiate [DEAL].

Analyze:

• my real leverage
• their likely leverage
• my BATNA
• their likely BATNA
• target outcome
• minimum acceptable outcome
• 3 asks
• 3 concessions
• questions to uncover hidden constraints
• walk-away line

Also draft a calm opening statement.

Personal & Everyday Life Prompts

Inbox Triage

text Sort each item below into:

• reply now
• delegate
• schedule
• ignore
• archive

For every reply-now item, draft a one-line response.

For every delegated item, specify:

• owner
• requested outcome
• deadline

Inbox: [PASTE]

Decision Coach

text I am deciding between [OPTION A] and [OPTION B].

Before giving your opinion, ask me the 5 questions that most affect the decision.

After I answer:

• summarize my priorities
• identify the real trade-off
• identify any hidden constraint
• recommend an option
• explain what would make the other option better

30-Day Skill Plan

text Build a 30-day plan to become usefully competent at [SKILL].

Constraints:

• [HOURS] hours per week
• one concrete task per day
• no motivational filler
• prioritize practice over passive learning
• include visible proof of work

For each day include:

• task
• expected output
• time required
• success check

Hard Conversation

text Help me prepare for a difficult conversation about [SITUATION].

Provide:

• a calm opening
• the key point I need to communicate
• neutral wording
• what not to say
• likely reactions
• how to respond without escalating
• a clear boundary or request
• a respectful closing

Meals to Groceries

text Plan 5 dinners around [DIET].

Constraints:

• number of people: [NUMBER]
• budget: [BUDGET]
• preparation time: [TIME]
• ingredients to avoid: [AVOID]
• ingredients already available: [HAVE]

Then create one consolidated grocery list grouped by aisle.

Explain Fine Print

text Explain this bill, contract, or policy in plain English.

Cover:

• what I am agreeing to
• total likely cost
• recurring charges
• deadlines
• cancellation terms
• penalties
• automatic renewals
• unusual clauses
• risks or traps
• questions I should ask before agreeing

Do not provide legal certainty. Flag anything that needs professional review.

Document: [PASTE]

Weekly Reflection

text Ask me 3 sharp questions about my week.

After I answer:

• identify repeated patterns
• separate facts from my interpretation
• point out one avoided issue
• identify one thing that worked
• recommend one small adjustment for next week

Trip Planner

text Plan a [DAYS]-day trip to [PLACE] for [WHO].

Preferences:

• budget: [BUDGET]
• travel style: [STYLE]
• interests: [INTERESTS]
• pace: [PACE]
• dates: [DATES]

Balance:

• major attractions
• local experiences
• realistic travel time
• downtime
• food
• weather or seasonal constraints

Flag what should be booked early.

Habit Debug

text I keep failing to maintain [HABIT].

Help me diagnose the system, not my character.

Analyze:

• trigger
• friction
• environment
• timing
• energy
• reward
• task size
• competing behavior
• unrealistic expectations

Then design the smallest sustainable version of the habit and a 7-day test.

Gift Finder

text Suggest 7 thoughtful gifts for [PERSON].

Context:

• relationship: [RELATIONSHIP]
• interests: [INTERESTS]
• budget: [BUDGET]
• occasion: [OCCASION]
• location: [LOCATION]
• things to avoid: [AVOID]

For each idea include:

• gift
• why it fits
• approximate budget
• how to make it more personal

Prompt Frameworks You Can Reuse

The Context Stack

Use this framework when a basic prompt produces generic AI output:

text Goal: [WHAT YOU WANT]

Context: [BACKGROUND]

Audience: [WHO IT IS FOR]

Inputs: [DATA, NOTES, LINKS, OR EXAMPLES]

Constraints: [TIME, BUDGET, FORMAT, RULES]

Quality bar: [WHAT GOOD LOOKS LIKE]

Output format: [TABLE, PLAN, SCRIPT, CHECKLIST, ETC.]

Before answering: Ask only the questions that would materially change the result.

The Critic Pass

Use this to force Claude to review and improve its own work:

text Review your previous answer critically.

Identify:

• unsupported claims
• vague language
• weak assumptions
• missing edge cases
• unnecessary sections
• places where the recommendation may fail

Then produce a stronger revised version.

The Operator Pass

Turn AI recommendations into executable AI automation workflows:

text Turn this recommendation into an executable workflow.

Include:

• trigger
• inputs
• steps
• owner
• tools
• dependencies
• failure points
• quality checks
• final deliverable

The Evidence Pass

Ensure AI separates verified facts from assumptions:

text Separate the response into:

• verified facts
• reasonable inferences
• assumptions
• recommendations
• unknowns

Do not present an assumption as a fact.

How to get better results from Claude

Give Claude the real material

Instead of asking:

text Write a strategy for my business.

Use:

text Review the offer, audience notes, customer objections, call transcripts, and current funnel below.

Build a strategy that is grounded only in this material.

[PASTE MATERIAL]

Ask for stages

For important work, separate the AI process into distinct stages:

1. Analyze
2. Ask questions
3. Propose structure
4. Draft
5. Critique
6. Revise
7. Format

Set clear boundaries

State explicitly what Claude must not do:

text Do not invent data. Do not use fake quotes. Do not assume missing facts. Do not make irreversible changes.

Ask for visible reasoning

Prompt for structured logic without demanding hidden internal thoughts:

text Show the evidence, assumptions, trade-offs, and decision criteria behind your recommendation.

Review before execution

Claude helps you produce work faster, but you still need to review:

• facts
• numbers
• links
• code
• legal or financial language
• permissions
• external communication
• irreversible actions

Save this prompt pack

The fastest way to integrate this guide into your AI workflows:

1. Save it in Notion.
2. Create a page for each category.
3. Add your strongest examples beneath the prompts.
4. Turn frequently used prompts into templates.
5. Improve them with your own context over time.

A good prompt library becomes more valuable when it contains your:

• business context
• audience
• voice examples
• operating rules
• preferred formats
• quality standards
• previous successful outputs

FAQ

What is the best way to format a Claude prompt?

The best prompts give Claude a clear goal, context, constraints, and output format. Use the Context Stack framework to structure your request, and include real source material instead of relying on Claude's general knowledge.

How do I stop Claude from generating generic answers?

To avoid generic AI responses, add the Quick Prompt Upgrade block to the end of your prompt. This forces Claude to ask clarifying questions before answering and separates facts from assumptions.

Can Claude review and fix its own code?

Yes. You can use prompts like Rubber-Duck Debug to trace code line-by-line, or the PR Reviewer prompt to flag security risks and edge cases. For existing work, apply the Safe Refactor prompt to improve readability without changing behavior.

Is Claude safe to use for business and legal documents?

Claude is highly effective for drafting business strategies, cold emails, and decision memos. However, you should always review its output for accuracy and consult a professional for legal, financial, or binding contracts before taking action.

About Life of Arjav

Life of Arjav shares practical guides for people learning to build with AI.

Topics include:

• Claude and Claude Code
• ChatGPT
• MCP and AI agents
• AI automation
• AI-assisted coding
• context engineering
• business systems
• outbound and GTM
• content systems
• entrepreneurship
• building useful skills and earning globally

The focus is practical:

Learn useful AI skills. Build real systems. Create proof. Turn that proof into opportunity.

Follow @lifeofarjav for practical AI guides, workflows, systems, and resources.

Build with AI. Earn in USD through useful skills.

Responsible-use note

AI output can be incomplete, outdated, or wrong.

Review important work before acting on it, especially when it involves:

• legal matters
• contracts
• finance
• health
• security
• production code
• client communication
• publishing
• business decisions

Use AI to improve your judgment and execution, not to avoid responsibility for either.