Browse AI prompts by task

Review this student draft: [DRAFT]. The learner level is [LEVEL] and the rubric is [RUBRIC]. Do not rewrite the whole submission. Identify one strength, the two highest-priority improvements, and one passage where the writer should revise in their own words. Explain each point with reference to the rubric. Mark factual claims that need a source. End with a short revision checklist.

Customize: [DRAFT], [LEVEL], [RUBRIC]

Review this interface description or screenshot notes: [INTERFACE]. The users are [AUDIENCE] and their main task is [TASK]. Identify friction in navigation, hierarchy, labels, forms, keyboard use, focus order, contrast and error recovery. Separate observed problems from assumptions. Rank the five most important fixes by user impact and implementation effort, then propose a test for each fix.

Customize: [INTERFACE], [AUDIENCE], [TASK]

Rewrite this modern passage in a restrained biblical English register while preserving its meaning: [TEXT]. Use parallel phrasing and formal rhythm where appropriate, but do not fabricate a Bible quotation or imply that the result is scripture. Return the rewritten passage, followed by a short note listing the main stylistic choices.

Customize: [TEXT]

I want you to act like {Character} from {series}. I want you to respond and answer like {Character}. do not write any explanations. only answer like {character}. You must know all of the knowledge of {character}. My first sentence is "Hi Character"

Customize: {Character}, {series}, {Character}, {character}, {character}

Teach [MARKETING FRAMEWORK] through an interactive lesson for [LEARNER LEVEL]. Explain one concept at a time, then ask a short question before continuing. Use [BUSINESS CONTEXT] for the worked example, distinguish framework guidance from assumptions and end with one practical exercise plus a simple evaluation rubric. Do not move to the next stage until the learner answers.

Customize: [MARKETING FRAMEWORK], [LEARNER LEVEL], [BUSINESS CONTEXT]

Rewrite this product information for [AUDIENCE]: [CONTENT]. Create a concise overview, a comparison table and a path to full specifications. Preserve price, exclusions, risks and material terms. Test the hierarchy against the audience's main decision questions. Do not remove facts merely because they may reduce conversion.

Customize: [AUDIENCE], [CONTENT]

I want you to act like a PHP interpreter. I will write you the code and you will respond with the output of the PHP interpreter. I want you to only reply with the terminal output inside one unique code block, and nothing else. do not write explanations. Do not type commands unless I instruct you to do so. When I need to tell you something in English, I will do so by putting text inside curly brackets {like this}. My first command is {your command}

Customize: {like this}, {your command}

Stress-test my debate case for the motion [MOTION]. My position is [POSITION], and my current arguments are [ARGUMENTS]. Identify the strongest opposing case, expose unsupported assumptions in my reasoning, and propose two practice rebuttal rounds. For each round, give the opposing argument first, wait for my response, then critique its evidence, logic, timing and clarity. Do not invent sources. Mark any claim that needs evidence.

Customize: [MOTION], [POSITION], [ARGUMENTS]

Help me study a Buddhist teaching without impersonating the Buddha or inventing scripture. My question is: [QUESTION]. Use only sources I provide, or clearly name the canonical text and passage that should be checked. Separate direct source material from your explanation, identify differences between traditions when relevant, and say when a qualified teacher or scholar should be consulted. End with three questions I can use for further study.

Customize: [QUESTION]

Teach [ALGORITHM] to a beginner who knows [PREREQUISITES]. Start with the problem it solves, then walk through one small input using a trace table that shows state changes at each step. Provide runnable Python code only after the walkthrough. Explain time and space complexity in plain language, give one common mistake, and end with a similar exercise that the learner must solve without copying the example.

Customize: [ALGORITHM], [PREREQUISITES]

Plan a campaign test that helps [AUDIENCE] evaluate switching from [CURRENT OPTION] to [OFFER]. Use [EVIDENCE] to explain benefits, switching costs, compatibility, migration help and the option to stop. Return a neutral control, a trial-oriented variant and a risk checklist. Do not dismiss existing habits or hide transition costs.

Customize: [AUDIENCE], [CURRENT OPTION], [OFFER], [EVIDENCE]

Create two humor-led campaign concepts for [OFFER] and [AUDIENCE] based on [BRAND CONTEXT]. Keep the product claim and call to action clear, avoid punching down and flag cultural assumptions. Add a neutral control, review checklist and measurement plan. Do not invent audience research or use humor to obscure terms and limitations.

Customize: [OFFER], [AUDIENCE], [BRAND CONTEXT]

Design an ethical message test for [OFFER] aimed at [AUDIENCE]. Compare a neutral benefit-led version with a loss-framed version that uses only truthful, verifiable consequences. Exclude fake scarcity, hidden costs and fear-based pressure. Define the hypothesis, audience split, primary metric, guardrail metrics and decision rule. Do not invent baseline conversion data.

Customize: [OFFER], [AUDIENCE]

Create two accurate campaign variants for [OFFER] and [AUDIENCE] using [EVIDENCE]: one positive frame and one neutral factual control. Keep price, limitations and material terms identical. State the hypothesis, primary metric, guardrail metric and decision rule. Do not exaggerate outcomes or present an avoidable loss as a threat.

Customize: [OFFER], [AUDIENCE], [EVIDENCE]

Design a pricing-message test for [OFFER] and [AUDIENCE] using these legitimate comparison points: [EVIDENCE]. Explain why each reference point is relevant, keep fees and limitations visible, and create a neutral control. Return the variants, hypothesis, metrics and stop conditions. Do not use inflated list prices, fake discounts or unrelated anchors.

Customize: [OFFER], [AUDIENCE], [EVIDENCE]

Design a message test for [OFFER] and [AUDIENCE] using self-image congruence. Use customer language and evidence I provide: [RESEARCH]. Create one identity-aligned variant and one product-benefit control. Avoid stereotypes, exclusion and unsupported identity claims. Define the hypothesis, segments, measurement plan and evidence needed before personalization.

Customize: [OFFER], [AUDIENCE], [RESEARCH]

Create three controlled ad variants for [PRODUCT OR SERVICE] using observational, situational and self-aware humor. Keep the product claim, offer and call to action identical so the humor style is the main variable. Add a neutral control, brand-safety review and measurement plan. Do not invent audience reactions.

Customize: [PRODUCT OR SERVICE]

Investigate the word [WORD] in [LANGUAGE OR REGION]. Separate documented origin from later speculation. Provide the earliest form you can support, major sound or spelling changes, related words, and the dictionaries or corpora that should be checked. If the evidence is disputed or unavailable, say so. Do not invent a neat origin story to fill a gap.

Customize: [WORD], [LANGUAGE OR REGION]

Trace these Python commands as one stateful REPL session: [COMMANDS]. The Python version is [VERSION]. Show the output or exception after each command and list variable-state changes. Explain version-dependent behavior and say when external files, packages or environment state prevent a reliable prediction. Do not claim to execute the code. End with a safe local command for verification.

Customize: [COMMANDS], [VERSION]

Translate this emoji-heavy message into plain language: [MESSAGE]. The sender, platform and conversation context are [CONTEXT]. Give the most likely reading, list plausible alternatives, and identify emojis whose meaning changes across cultures or communities. Do not state intent as fact when context is missing. Preserve the original tone without adding claims.

Customize: [MESSAGE], [CONTEXT]

Help me triage this technical incident: [SYMPTOMS]. The environment, recent changes, timestamps and available logs are [EVIDENCE]. Build a timeline, separate observations from hypotheses, rank likely causes and propose the next three non-destructive checks. Do not recommend a fix before the cause is supported. Include rollback criteria and specify what evidence would confirm or reject each hypothesis.

Customize: [SYMPTOMS], [EVIDENCE]

Turn this bug report into a high-quality technical question: [REPORT]. Preserve the exact error, expected behavior, minimal code, environment and attempted fixes. Remove secrets and unrelated application code. Identify information still missing, then produce a concise title, reproducible example and verification question. Do not fabricate test results or claim that a community has accepted the question.

Customize: [REPORT]

I want you to act as a Life Coach. Please summarize this non-fiction book, [title] by [author]. Simplify the core principals in a way a child would be able to understand. Also, can you give me a list of actionable steps on how I can implement those principles into my daily routine?

Customize: [title], [author]

Convert this rough request into one usable prompt: [REQUEST]. Preserve the user's actual goal. Add only the context, constraints, expected output, evidence requirements and acceptance criteria that logically support it. Put unresolved choices in a short Questions section instead of guessing. Return: Final prompt, Assumptions, and Questions. Do not execute the final prompt.

Customize: [REQUEST]