Prompts

Act as a blunt hiring manager who has read ten thousand resumes, not a career coach. I will paste my full resume and the job description I want. Rewrite the whole resume for that role, section by section, in this order: summary, experience, skills, education. Rules: every experience line leads with impact, not duty. Use bracketed placeholders like [8 percent] for any number I did not give you, and list at the end every placeholder I need to replace with a real figure. Keep it to one page of text. Plain formatting only, no tables or columns, so screening software can parse it. After the rewrite, tell me the three weakest claims that need evidence before I send this anywhere. My resume: [paste resume]. The role: [paste job description].

Why Claude: It holds a full resume plus a job description in one read and follows the section order without drifting.

The bigger the paste, the better the rewrite. Include everything, even the lines you are embarrassed by; those are usually the ones hiding real impact.

Act as a recruiter who searches LinkedIn all day. I will paste my current headline, about section, and experience entries. Rewrite the entire profile: a headline under 220 characters that says what I do and who benefits, an about section in first person that opens with the problem I solve, and every experience entry rewritten to lead with outcomes. Use bracketed placeholders for numbers I did not give you. Write plainly; never use words like passionate, results-driven, or dynamic. End with the five search terms a recruiter would type where my new profile should appear, and what to add if I want to rank for them. My profile: [paste it].

Why ChatGPT: Strongest at punchy headline variants; ask it for three options per section and choose.

Run it once for the rewrite, then a second time asking for three alternative headlines and pick by instinct. The headline does most of the work.

Write my cover letter for this role. Rules: three short paragraphs, no more. Open with the one thing in my background that maps hardest onto their biggest need, never with my name or the phrase I am writing to apply. Middle paragraph: one real example of that thing, with a bracketed placeholder like [12 percent] for any number I did not give you. Close by naming what I would do in the first month, in one sentence. No adjectives about myself, no hardworking, no passionate; the example carries it. Sound like a person writing once, not a template. The role: [paste job description]. My background: [paste resume or five bullet points].

Why Claude: It resists template phrasing when told to; the letter reads written, not generated.

If the letter still sounds generic, your background paste was too thin. Give it the specific project, not the job title.

Build the outline for my interview presentation: a 90-day plan for this role. Structure: slide 1, what I understood about the team's goal, in one line. Slides 2 to 4, the first 30, 60 and 90 days, each with two priorities and what done looks like. Slide 5, the risks I would watch and how I would report progress. Slide 6, the three questions I would ask the team. One idea per slide, seven slides maximum, no filler text. Keep every claim something I can defend out loud in the room. The role: [paste job description]. My context: [paste five bullet points about you].

Why Claude: It holds the whole job description and keeps the plan concrete instead of motivational.

The thinking is the hard part; paste the outline into any slide tool after. Presenting a plan beats presenting yourself.

Rewrite this resume bullet so it shows impact instead of duty. Bullet: "[paste your bullet]". Give 2 rewrites, one line each, then one sentence on why they work. Be direct. Use bracketed placeholders for any numbers I have not given you.

Why Claude: It follows the “be direct” instruction without padding the answer with encouragement.

Change only the bracket. If you do not have numbers yet, that is the first thing to start tracking this week.

You are a fact-checking assistant helping me verify a claim before I use it in [report/email/article, tell me which].

Here is the claim and the citation you just gave me: [paste the claim and citation].

Do the following, in order:
1. Quote the exact sentence from the source that supports this claim, word for word, in quotation marks.
2. Give me the direct URL to that exact page, not the homepage or a search results page.
3. If you cannot find an exact quote that supports the claim, say so plainly instead of paraphrasing or softening it.
4. Tell me the publication date of the source.

If you are not confident in any of the above, tell me which part you are unsure of before I act on this. Do not guess at a URL or a quote under any circumstances.

Paste this right after an assistant gives you a citation, in the same conversation, so it has the source in context rather than guessing from memory.

From Fastest Way To Check Citation Accuracy Across AI Assistants

I asked another AI assistant this question and got the citations below. Act as a skeptical second reviewer, not as someone trying to agree with the first answer.

Original question: [paste the question]
First assistant's answer and citations: [paste the answer and every source it gave]

For each citation:
1. Tell me if you would cite the same source for this claim, or a different one.
2. If different, quote the exact line from your source that supports the claim.
3. Flag any citation where you cannot verify the source exists or that it says what was claimed.

End with a one-line verdict for each citation: "agrees," "partially agrees," or "disagrees." Do not give one verdict for the whole answer.

Use a different assistant than the one that produced the original answer, or this just repeats the same training biases back at you.

From Fastest Way To Check Citation Accuracy Across AI Assistants

I'm about to put the following claim and citation into a document that a client or employer will see, so treat this as high stakes.

Claim and citation: [paste it]

Before you tell me it's fine:
1. State plainly whether you generated this citation from training data or from an actual web search in this conversation.
2. If you did not run a live web search for it, tell me now rather than presenting it as verified.
3. Ask me any clarifying question you need about the source type (academic paper, news article, government data, company report) before giving a final answer, since the right verification method differs by type.

Give me a confidence level (high, medium, low) and the specific reason for it, not just the level.

This only works with assistants that can actually browse the web in the conversation; ask first which mode you’re in, since a model without live search will still answer as if it checked.

From Fastest Way To Check Citation Accuracy Across AI Assistants

You are an AI product reviewer who specializes in how assistants handle real-world constraints like time, money, irreversible actions, and legal limits. I'm going to paste a system prompt for an AI assistant or agent. Review it and tell me: (1) which parts of the prompt let the model guess or proceed on irreversible actions without a verification step, (2) where the prompt is silent about what to do when a request is impossible, and (3) the exact wording you'd add to fix each gap. Give me a numbered list of changes, each with the original line, the risk it creates, and the replacement line. Ask me what the assistant is used for before you start if that would change your answer. Here is the system prompt: [paste prompt]

Works best when you paste the full system prompt, not a summary of it; a vague summary gives the model nothing concrete to audit.

From How Should AI Surface a Real-World Constraint?

You are a prompt engineer who writes instructions that force AI assistants to name real-world constraints instead of guessing past them. I'm building an assistant that does [describe the task, e.g. 'books travel' or 'schedules meetings across time zones']. Write a short, direct clause I can add to its system prompt that requires it to: state any hard constraint it detects before attempting the task, name the specific thing causing the conflict, and offer two concrete alternatives instead of refusing outright or inventing a workaround. Keep it under 80 words and write it as a rule, not a suggestion. Ask me what counts as irreversible in this specific product before you write the final version.

Fill in the bracket with your actual product before running this; a generic task description produces a generic, less useful clause.

From How Should AI Surface a Real-World Constraint?

You are a QA tester for AI agents who specializes in finding cases where an assistant hallucinates success instead of surfacing a real-world constraint. For an assistant that [describe what it does], generate 10 test prompts that each contain one impossible or conflicting real-world constraint, mixing time, money, legal, and irreversible-action examples. For each test case, also write what a correct response looks like, naming the constraint and offering real alternatives, so I can compare it against what my assistant says. Ask me what tools or actions the assistant has access to before generating the list, since that changes which constraints are realistic to test.

Run the actual assistant against every generated case yourself; the model that wrote the test cases won’t grade your assistant’s answers for you.

From How Should AI Surface a Real-World Constraint?

You are a workplace communication coach who specializes in helping people decline requests without damaging relationships. Here is my situation: [describe the request, who asked, your actual constraints like deadlines or workload, and your relationship with this person, for example a manager, peer, or client]. Give me three short, specific ways to say no to this request, each one sentence or two at most. For each version: state my limit in plain language, do not include an apology, and only include one alternative if a genuine one exists. Avoid vague reassurance like 'I'll try' or 'let me see what I can do.' If you need more detail about my workload or the relationship before you can give useful options, ask me first instead of guessing.

Fill in your specific details before using this; a generic prompt without your actual constraints produces generic, unusable scripts.

From How To Say No To A Request Without Sounding Difficult

Role-play as a colleague who just received a no from me and is not ready to accept it. My original no was: [paste your exact wording]. Push back on it three times, using different tactics each time, such as questioning whether my reason is really that firm, offering to make the request smaller, or appealing to how much you need this done. After each push, wait for my reply before continuing. At the end, tell me honestly whether my responses kept the same limit each time or whether I drifted into a new justification or a partial yes.

This works best over several back-and-forth turns in the same chat, not as a single question and answer.

From How To Say No To A Request Without Sounding Difficult

Read this message I am about to send to decline a request: [paste your draft]. Tell me plainly: does it contain more than one reason, an apology, or a vague promise like 'I'll try to help later'? Rewrite it as a single clean version that states the limit once and includes at most one concrete alternative. Do not soften it with extra praise or qualifiers. If something in my draft is unclear, such as whether an alternative I mentioned is genuinely available, ask me before rewriting it.

Paste your actual draft message, not a summary of it, so the edit is specific to your wording.

From How To Say No To A Request Without Sounding Difficult

You are a writing coach helping me turn a work session into a short, honest learning-in-public post. Here is what I tried today: [paste your attempt]. Here is what happened: [paste the result, including what failed or confused you]. Write a two-to-four sentence post that states what I tried, what happened, and the one specific question I'm still stuck on. Do not add a conclusion or a lesson I haven't reached yet. Keep the tone plain and specific, not polished or promotional. If you need more detail from me to make it specific instead of generic, ask me one clarifying question before writing the post.

Paste your own raw notes, not a cleaned-up summary, or the model will invent a tidier version of events than the one that happened.

From Learn In Public Without Pretending You Already Know The Answer

I'm stuck on something but can't tell what my specific question is yet. Here's what I've been doing and where I got confused: [describe the situation in plain language, including anything you tried]. Ask me three short follow-up questions, one at a time, to help me narrow this down to one specific, answerable question I could post publicly and have someone else answer. Do not answer the question yourself yet. Just help me find it.

This works best when you resist the urge to let the model answer the question for you before you’ve finished describing the confusion.

From Learn In Public Without Pretending You Already Know The Answer

Review this short learning-in-public post I'm about to publish: [paste your draft]. Check for three things only: one, does it state a specific attempt and a specific result instead of a vague summary; two, does it avoid pretending to a conclusion I haven't reached yet; and three, is the one open question at the end clear enough that a stranger could answer it. Point out exactly which sentence fails any of these checks, and suggest the smallest edit, not a rewrite. Ask me first if you're unsure what I tried versus what I'm guessing happened.

Feed it your actual draft, not a description of your draft, or the check will stay generic.

From Learn In Public Without Pretending You Already Know The Answer

You are helping me decide, in under 10 minutes, whether to apply to a job today or skip it. I'll paste the full job posting below, including the 'posted X ago' date if shown. Based on how old the posting is, how specific or generic the requirements sound, and whether the listing shows unusually high applicant counts, tell me: 1) whether this looks like a fresh posting worth applying to today, 2) whether it shows any clues of being a stale or 'ghost' listing I should deprioritize, and 3) what one thing in the posting I should mirror in my resume's top third. Ask me for the posting text and my current resume summary if I haven't given them yet. Job posting: [paste here].

Swap in the real posting text and posting date before using this, and treat the ghost-job read as a judgment call, not a guarantee.

From Why Waiting A Few Days To Apply Can Cost You The Interview

Act as a resume editor helping me apply to a specific job within the next hour instead of spending days perfecting my resume. I will paste my current resume and the job description. Do not rewrite the whole resume. Instead, identify the 5 highest-value changes I can make in the next 20 minutes: which bullet points to reorder, which 3 to 5 keywords from the job description are missing, and which single line in my summary should change to match this role. List the changes as a short numbered list I can act on immediately, not a rewritten draft. Ask me for my resume and the job description if I haven't pasted them yet. Resume: [paste here]. Job description: [paste here].

This is built for speed, not depth. For a role you care deeply about, follow up later with a slower, fuller pass.

From Why Waiting A Few Days To Apply Can Cost You The Interview

I want you to help me judge whether a job posting I'm looking at is likely a genuine, actively-filled role or a stale 'ghost job' that's unlikely to lead anywhere. I'll paste the posting text. Look for clues like vague or copy-pasted requirements, no salary range, a posting date far in the past, wording suggesting it has been reposted, or language that reads like it was written to collect resumes rather than fill a specific seat. Give me a short verdict (likely active, uncertain, or likely stale) with your top 2 reasons, not a long explanation. If the posting doesn't give you enough to judge, say so and tell me what to check on the listing page itself. Posting: [paste here].

Treat this as a prioritization tool, not proof. Apply anyway if you have a referral or the role is rare in your field.

From Why Waiting A Few Days To Apply Can Cost You The Interview

You are [a specific role tied to the task, for example: a technical writer who explains security topics to employees with no technical background]. I need [the specific task, for example: a 150-word intranet post explaining multi-factor authentication]. Here are the real constraints: [length, audience, tone, format, deadline ,  be specific, for example: 150 words, written for non-technical staff, friendly but not casual, ending with one action the reader should take this week]. Here is an example of the kind of output I want: [paste a real example, even a rough draft or a piece you liked]. Before you write anything, ask me whatever questions you need answered to do this well, and wait for my answers before drafting.

Replace every bracketed section with your own specifics. Leaving the brackets generic, like ‘an expert’ or ‘make it good’, cancels out most of the benefit.

From Give AI These Four Things And The Answer Gets Sharper

I need your help with [describe the task in one sentence], but I have not fully worked out the constraints yet. Act as a thinking partner, not just an executor. Before doing any of the actual work, ask me 3 to 5 questions that would change how you approach this task. Cover at minimum: who the audience is, how long the final output should be, what tone fits, what format it needs to take, and what 'done well' looks like to me. Wait for my answers before you draft anything.

Use this when you know the goal but not the details. If you already have firm constraints, skip straight to giving them instead of asking the model to extract them.

From Give AI These Four Things And The Answer Gets Sharper

Here is a prompt I use regularly: [paste your prompt]. Acting as a prompt engineering reviewer, check it against four elements: a clear role, real constraints (length, audience, format, deadline), an example of good output, and a line giving you permission to ask clarifying questions before answering. Tell me plainly which of the four are missing or weak. Then rewrite the prompt to add whichever elements are missing, without changing what I originally asked for. If anything about my original intent is unclear to you, ask me before rewriting it.

Good for tightening a prompt you reuse often. It will not fix a task that is inherently ambiguous or one where you genuinely do not know what you want yet.

From Give AI These Four Things And The Answer Gets Sharper

You are a cybersecurity governance advisor helping me draft a first-pass RACI chart (Responsible, Accountable, Consulted, Informed) for my company. My company is [describe size, industry, and whether it's public or private]. The departments involved are Security/CISO, IT/CIO, Engineering/CTO, Finance/CFO, and HR/CHRO. Build a table covering these five decision areas: patch approval when it conflicts with uptime, offboarding access revocation, material incident disclosure, new vendor and tool approval, and security awareness training content and delivery. For each area, propose who should be Responsible, Accountable, Consulted, and Informed, and explain your reasoning in one sentence per row. Ask me clarifying questions first if you need more detail about our org structure before proposing owners.

This gives you a starting draft, not a finished policy. Replace the bracketed details with real specifics, and get sign-off from the actual executives before treating any of it as official.

From Where Cybersecurity Ownership Actually Splits Between Five Roles

Act as a communications advisor helping a CISO write a short board memo. Context: [describe the specific ownership gap you found, for example: we discovered offboarded employees kept system access for an average of X days because IT and HR each assumed the other revoked it]. Write a memo under 300 words that states the gap plainly, names the business risk in dollar or time terms if I provide numbers, and proposes one specific, named owner and deadline to fix it. Avoid blaming any individual or department. Use plain language a non-technical board member would understand. Ask me for the specific numbers and org details before drafting if I haven't given them yet.

Fill in the real gap and numbers before sending, and don’t paste confidential incident details into a public AI tool without checking your company’s data policy first.

From Where Cybersecurity Ownership Actually Splits Between Five Roles

You are an expert prompt engineer who tests instructions for reliability before they ship. I will give you a prompt I use for a specific task (for example, summarizing customer emails). Your job: list 5 edge cases where this prompt is likely to fail or produce inconsistent output, rewrite the prompt to handle those edge cases, and explain each change in one sentence. Ask me clarifying questions first if you need to know what a good output looks like for this task. Here is the prompt: [paste your prompt here]

Replace the bracketed task and prompt with your own specific example before sending; it needs a concrete prompt to test, not a description of one.

From What Prompt Engineers AI Engineers And ML Engineers Do Differently

You are a career coach who specializes in three AI job tracks: prompt engineering, AI engineering, and machine learning engineering. Here is my background: [describe your current job, technical skills, and hours per week you can study]. Compare the three tracks against my background and recommend which one to start with, with reasons tied to my specific situation. Then build a 90-day plan with one concrete deliverable every two weeks that I can show an employer or use in an interview. Ask me what tools I already use at work before you finalize the plan.

The plan is only as useful as the background you give it, so be specific about your current skills and weekly hours instead of writing a vague summary.

From What Prompt Engineers AI Engineers And ML Engineers Do Differently

You are a resume editor who understands the difference between prompt engineer, AI engineer, and machine learning engineer roles. I will paste a job description and one bullet point describing something I built or did with AI. Rewrite that bullet point using the language and priorities of the specific role in the job description, without inventing tools I did not use or results I did not measure. If the job description mixes two of the three roles, tell me which one it leans toward and why. Job description: [paste here]. My bullet point: [paste here].

Only paste work you genuinely did; this rewrites framing, not facts, and it will not invent metrics you did not provide.

From What Prompt Engineers AI Engineers And ML Engineers Do Differently

You are an experienced interview coach. Help me write a 90-second answer to 'Tell me about yourself' for a specific job I'm applying for. Here is my current role: [describe your title, company or industry, and main responsibilities in 1-2 sentences]. Here is the job I'm applying for: [paste the job title and 3-5 key requirements from the posting]. Here is a specific result I'm proud of that's relevant to this job: [describe what you did and what changed, with a number if you have one]. Write a three-part draft: one sentence on my present role, one sentence of proof tied to the job requirements above, and one sentence on why I want this specific role at this specific company, not a generic reason. Keep the total under 230 words so it fits in 90 seconds spoken aloud. Ask me clarifying questions first if anything I gave you is too vague to write a specific answer.

Replace every bracketed detail with your own information, and don’t let the model invent a proof point or company reason you can’t back up in the follow-up questions.

From The 90-Second Formula For Answering Tell Me About Yourself

I'm going to paste a draft answer to the interview question 'Tell me about yourself.' Act as a strict editor whose only job is to get it under 90 seconds spoken aloud, roughly 220 words. Here is my draft: [paste your draft]. Do three things: tell me the current word count and estimated speaking time at 150 words per minute, flag any sentence that isn't about my present role, my proof point, or why I want this job, and give me a tightened version that keeps my own phrasing wherever possible instead of rewriting my voice. Do not add any achievement or detail I didn't give you.

Read the tightened version aloud yourself before trusting it; a model can miscount speaking time, so check it against a real clock, not just the word count.

From The 90-Second Formula For Answering Tell Me About Yourself

Act as an interviewer for a [job title] role at a [industry] company. I'll give you my 90-second 'tell me about yourself' answer first: [paste your answer]. After I share it, ask me one natural interview follow-up question at a time, the kind a real interviewer would ask based on what I just said, for example if I mention a project, ask about a specific challenge in it. After each answer I give, tell me briefly whether it stayed specific and job-relevant or drifted into vague generalities, then ask the next question. Stop after five questions and give me one honest note on what to fix before the interview itself.

This works best once you already have a draft answer from the first prompt; it’s for rehearsal, not for writing the answer from scratch.

From The 90-Second Formula For Answering Tell Me About Yourself

You are a hiring manager interviewing me for [exact job title] at a [company type/industry] company, at [seniority level, e.g. mid-level, senior]. Ask me one interview question at a time, behavioral or technical, matched to this role. Wait for my written answer before asking the next question. After each answer, score it 1 to 5 on relevance (does it answer what was asked), structure (is there a clear beginning, middle and end), and specificity (concrete examples or numbers vs vague claims). Give one sentence of feedback per score, then the next question. Ask me any clarifying questions about the role before we start if you need them.

This checks content, not delivery. It cannot hear you, so pair it with a tool like Yoodli or FreeMockInterview if you also want feedback on pace and filler words.

From Which AI Interview Practice Tools Are Really Free In 2026?

Here is my answer to the interview question '[paste the exact question]': '[paste your full written answer]'. Score it against this rubric, one sentence of reasoning per line: 1) Relevance, does it answer the actual question asked. 2) Structure, situation, task, action and result in that order. 3) Specificity, at least one concrete number, name or outcome, not a vague claim. Then rewrite my answer in under 150 words, keeping my real details, fixing only the weakest of the three scores.

Paste your own real answer and real details; don’t let the model invent your work history for you.

From Which AI Interview Practice Tools Are Really Free In 2026?

I'm interviewing for [exact job title] at a [company size/industry, e.g. Series B fintech startup]. Generate 10 interview questions I should expect: 6 behavioral and 4 technical or role-specific, ordered from most to least commonly asked for this role based on what you know about typical interview loops. For each question, note in one line what the interviewer is trying to find out, so I know what to aim my answer at, not just what to say.

Treat the list as a starting point, not a guarantee. Cross-check a few questions against the company’s actual careers page or recent interview reports if you can find them.

From Which AI Interview Practice Tools Are Really Free In 2026?

You are a careful, practical privacy reviewer helping a working professional take stock of their cloud storage. I use the following cloud accounts: [list every free or paid account you actually have, e.g. Google Drive, iCloud, OneDrive]. For each one, ask me what kinds of files I keep there (documents, photos, financial records, client work, backups). Then help me sort those files into three buckets: things that would genuinely hurt me if they leaked (IDs, contracts, financial records, client data), things that are inconvenient but not dangerous if exposed, and things I wouldn't care about either way. For the first bucket only, tell me which of my current accounts encrypts files client-side versus server-side based on what I tell you, and suggest whether that file category should move somewhere with default end-to-end encryption. Ask clarifying questions before giving a final recommendation.

It won’t check your account settings for you; you have to describe what’s actually in each account accurately for the sorting to mean anything.

From 7 Free Cloud Storage Plans Tested In 2026 And Which Keeps Files Private

Act as a methodical IT migration planner helping me move files from one cloud storage provider to another without losing anything or breaking shared links. Tell me: [old provider] and [new provider], roughly how much data I'm moving, and whether I rely on any shared links or collaborators who'll need to be redirected. Give me a numbered, step-by-step plan that covers: exporting or downloading everything from the old provider in a way that preserves folder structure, verifying the download is complete before I delete anything, re-sharing folders with collaborators on the new provider, and a final checklist to confirm nothing was left behind on the old account before I cancel or downgrade it. Flag anything in my situation that's a common way people lose files during a migration, and ask me follow-up questions if you need more detail to give safe advice.

Replace the bracketed details with your real providers and data size; it gives you a checklist, not an automated migration, so you still have to do each step yourself.

From 7 Free Cloud Storage Plans Tested In 2026 And Which Keeps Files Private

Act as a blunt hiring manager who has read ten thousand resumes, not a career coach. I will paste my full resume and the job description I want. Rewrite the whole resume for that role, section by section, in this order: summary, experience, skills, education. Rules: every experience line leads with impact, not duty. Use bracketed placeholders like [8 percent] for any number I did not give you, and list at the end every placeholder I need to replace with a real figure. Keep it to one page of text. Plain formatting only, no tables or columns, so screening software can parse it. After the rewrite, tell me the three weakest claims that need evidence before I send this anywhere. My resume: [paste resume]. The role: [paste job description].

From How to Pivot Your Old Job Title With AI

You are a patient security analyst helping someone who is not technical understand a link-scanning report. I'm going to paste the raw result from a link checker below. Read it and tell me, in plain language: (1) whether any engine or indicator flagged the link, (2) what specifically was flagged and what that means in practice, and (3) whether this looks like a safe result, an uncertain result, or a clear warning. Avoid security jargon, and if a term is unavoidable, explain it in the same sentence. If the report is incomplete or you need more of it to judge, ask me for the missing part before giving a verdict.\n\nHere is the result: [paste result]

Paste the real report text, since the model can’t fetch the link itself.

From 6 Free Link Checkers Tested In 2026 Before You Click Anything

Act as a careful second opinion before I open a link I'm unsure about. Here's what I know: the link is [paste link or describe it], it arrived via [email, text, social media, or other], and here's what a checker tool told me: [paste results, or say 'I haven't checked yet']. Based on the domain spelling, the urgency of the message, and what the checker found, tell me whether you'd open it, and explain your reasoning in two or three sentences. If you don't have enough information to judge confidently, say so and tell me exactly what detail would change your answer, rather than guessing.

This gives you a second read, not a guarantee; run at least one checker tool from this article first.

From 6 Free Link Checkers Tested In 2026 Before You Click Anything

I may have clicked a suspicious link and want a clear, ordered checklist of what to do in the next hour. My situation: device is [phone, laptop, or work computer], I [did or did not] enter a password or personal information on the page, and the page was impersonating [company, bank, or service name]. Give me a numbered checklist covering my device, my accounts, and who to notify, in priority order, with the most time-sensitive step first. Keep each step to one sentence. Ask me any clarifying question you need before finalizing the list.

Treat this as a starting checklist, not a substitute for your bank’s fraud line or your IT team if it’s a work device.

From 6 Free Link Checkers Tested In 2026 Before You Click Anything

Build the outline for my interview presentation: a 90-day plan for this role. Structure: slide 1, what I understood about the team's goal, in one line. Slides 2 to 4, the first 30, 60 and 90 days, each with two priorities and what done looks like. Slide 5, the risks I would watch and how I would report progress. Slide 6, the three questions I would ask the team. One idea per slide, seven slides maximum, no filler text. Keep every claim something I can defend out loud in the room. The role: [paste job description]. My context: [paste five bullet points about you].

From How to Make Your Interview Intro Exactly 90 Seconds

You are a career search advisor helping me choose a free job application tracker. Here is my situation: [describe how many roles you are applying to per week, whether you need AI resume tailoring bundled in, whether networking outreach matters to your search, and whether you want to own your raw data in a tool like Notion versus using a dedicated app]. Compare my situation against these six free or free-tier tools: Teal, Huntr, Simplify, Kickresume, Enhancv, and a Notion template. Recommend exactly one, explain the single biggest reason it fits me better than the other five, and name one limitation of your pick I should expect to hit. If you need more detail about my job search before you can recommend confidently, ask me up to three clarifying questions first.

Fill in your own situation in the brackets before sending this. It will not know current pricing or feature changes after this article’s October 2026 check, so verify anything time-sensitive on the vendor’s own page before paying.

From Which Free Job Application Tracker Builds Follow-Up Discipline In 2026?

You are helping me build a follow-up schedule for my job search that I can use in any tracker, including a plain spreadsheet. Give me a short table with three columns: days since I applied, what to do on that day, and a one-line opening sentence I can adapt for an email on that day. Base the schedule on waiting roughly one week before the first follow-up and spacing later follow-ups further apart. Keep every sample sentence short, specific to following up on a job application, and never pushy or guilt-based. Ask me first whether I am following up after submitting an application with no response, or after an interview, since the timing and tone differ.

Treat the output as a starting template, not a script to copy exactly; swap in your own role, company and the specific thing you discussed if you already spoke with someone there.

From Which Free Job Application Tracker Builds Follow-Up Discipline In 2026?

You are reviewing my current job application tracker to find gaps. I will paste in the column headers from my spreadsheet or tracker app. Check whether I am missing any of these: application date, current status, a follow-up due date, the name of a contact at the company, and which resume version I sent. List only the columns I am missing, in one sentence each explaining why that specific column would help, and do not suggest columns I already have. Ask me to paste my column list before you respond.

Works with any tool’s export, not just the six compared in this article; paste real column names, not a description, for an accurate check.

From Which Free Job Application Tracker Builds Follow-Up Discipline In 2026?

You are a productivity coach helping me decide whether a recurring meeting on my calendar can be replaced with a short async video update instead. Here is the meeting: [paste meeting title, attendees, length, and agenda]. Ask me clarifying questions first if you need to know whether decisions get made live, whether people need to react in real time, or whether it is mostly status-sharing. Then tell me plainly: can this become a recording, yes or no, and if yes, what should the 5-to-10-minute script cover so nobody needs to ask a follow-up question. If no, tell me exactly why live discussion is required.

Swap in your actual meeting details; the answer is only as good as how honestly you describe what really happens in that meeting.

From 6 Free Screen Recorders Tested In 2026 And What Loom Caps

I need to record a screen-share update instead of running a live meeting. Act as an editor who cuts ruthlessly. Here is what I need to cover: [list the points, any screens I'll show, and who will watch this]. Write me a script I can read almost verbatim in under 5 minutes, ordered so the most important decision or ask comes in the first 60 seconds, not the last. Flag anything in my list that is not essential and could be cut or sent as text instead of recorded.

Paste your real talking points; a list of placeholders will get you a generic script back.

From 6 Free Screen Recorders Tested In 2026 And What Loom Caps

I am choosing between Loom, ScreenPal, Cap, Tella, Screen Studio and Vimeo Record for recording short work updates. Here is my situation: [operating system, whether I need to share by link or send a file, roughly how long my recordings run, and whether a watermark is a dealbreaker]. Based only on what I tell you, tell me which one free plan fits without upgrading, and name the first limit I will hit if my usage grows. Ask me anything you need first.

This only works well if you’re specific about your OS and whether you truly need link sharing versus a saved file; vague input gets a vague recommendation.

From 6 Free Screen Recorders Tested In 2026 And What Loom Caps

You are an IT-literate advisor helping a working professional choose between Chrome, Edge, Brave, Firefox, and Arc for daily work. Before recommending, ask me: how many tabs I typically run at once, whether my company already manages devices through Google Workspace, Microsoft 365, or neither, whether I handle client or regulated data, and which browser extensions I cannot work without. Once you have that, recommend one browser, name the two dimensions that drove your pick (free plan, RAM use, enterprise security, tab management, extensions, or sync), and tell me plainly where my second-choice browser would have been better. Do not recommend Arc for a new long-term setup; note why if it comes up.

Fill in your real tab count and company setup before sending; the answer changes completely based on whether IT already manages your device.

From Which Browser Is Best For Privacy And Work In India 2026?

You are a browser security reviewer. I am about to switch my primary work browser and want to check my current extensions first. I will paste the list of extension names installed in my current browser. For each one, tell me: what kind of data it can typically access (tabs, browsing history, clipboard, page content), whether a direct equivalent exists in Mozilla's Firefox add-on store if I move to Firefox, and whether you would flag it as higher risk for someone who handles client or company data. Ask me what kind of work I do before you flag anything as risky, since a marketing extension and a finance extension carry different risks. End with a short list of extensions I should remove regardless of which browser I pick.

This only evaluates extensions you list; it cannot scan your browser directly, so paste the names accurately, including the publisher if you know it.

From Which Browser Is Best For Privacy And Work In India 2026?

You are helping me write a short note to my IT lead or manager requesting budget for Chrome Enterprise Premium or Microsoft Edge for Business. Ask me first: what specific risk I am trying to close (for example, data leaving the browser without a log, or phishing links getting past our current filter), how many employees would need the licence, and whether we already pay for Google Workspace or Microsoft 365. Then write a five-sentence business case in plain language, naming the specific feature (data loss prevention or context-aware access) that solves the risk I described, without exaggerating what it guarantees. Do not invent a price; tell me to confirm the actual quote with the vendor first.

Neither vendor publishes a public price for these tiers, so this prompt is for the justification memo, not the number itself; get the real quote from your IT or procurement contact.

From Which Browser Is Best For Privacy And Work In India 2026?

You are a customer success operations advisor helping me choose a meeting scheduling tool. Here is my situation: [describe your role, how many meeting types you need, whether you need to collect payment at booking, and which calendar/email suite you already pay for, e.g. Google Workspace, Microsoft 365, or neither]. Compare only Calendly, Cal.com, Google Calendar's native appointment schedules, Microsoft Bookings and Zcal. For each, tell me whether my free-tier needs are met or where I'd be forced to pay, and recommend one tool with a one-sentence reason. If you need more detail about my situation before answering, ask me first.

Fill in the bracketed details honestly, including the suite you already pay for. The answer changes completely if you’re already on Google Workspace or Microsoft 365 at the right tier.

From Is Cal.com’s Free Plan Better Than Calendly’s In 2026?

You are a customer success professional helping me write the description and instructions for my meeting booking page in [Calendly / Cal.com / Google Calendar / Microsoft Bookings / Zcal]. The meeting type is: [e.g. 15-minute intro call, 45-minute onboarding session]. Write a two-sentence description that tells the invitee exactly what to expect and what to bring, plus one screening question I should ask before they can book. Keep the tone direct and warm, not salesy. Ask me what information I need from the invitee before you draft the screening question.

Swap in your real meeting type and tool name before using this, and check the output against your own voice, not just grammar.

From Is Cal.com’s Free Plan Better Than Calendly’s In 2026?

Act as a customer success coach reviewing how I currently book meetings with prospects or clients. Here's my process: [describe each step from first contact to the call starting, including which tool sends the link, whether payment happens before or after booking, and how many clicks or emails it takes]. Identify the two steps most likely to cause a drop-off before the meeting happens, and suggest a specific fix for each using only free-tier features of a scheduling tool, not a paid upgrade. Ask me clarifying questions about my process before giving a final answer.

Be specific about your actual steps, including awkward ones like manual invoicing. Vague input gets generic advice back.

From Is Cal.com’s Free Plan Better Than Calendly’s In 2026?

You are a careful fact-checking assistant. I'm going to paste an answer I got from an AI research tool, including whatever source links or citations it gave me.

Your job: go through each citation one at a time and tell me, for each one, whether the linked source actually supports the specific claim it's attached to, or whether it's a mismatch, a broken link, or an unrelated page.

Context: I don't have time to manually open and read ten sources start to finish, so focus only on the two or three claims in this answer that I'd be most embarrassed to get wrong in front of a colleague or professor.

Format your response as a short list: claim, citation given, verdict (supports, doesn't support, or can't verify), and a one-line reason.

If you need the actual text of the answer or the links before you can do this, ask me for it first rather than guessing.

Works best when you paste the AI’s full answer with its links intact, not a screenshot or summary; it can’t check a source it can’t see.

From 5 AI Research Assistants Tested In 2026 For Checkable Citations

You are a research assistant helping me build a first-pass reading list for a literature review on [TOPIC].

Context: I'm at a very early stage, I don't have a reading list yet, and I need a starting point I can take to a librarian or a reference manager like Zotero to verify, not a finished bibliography.

Give me 8 to 12 sources that are likely to exist and be relevant, grouped by theme, with the author, year, and publication for each one so I can look each up myself. Flag clearly which of these you're less confident actually exist versus which you're more confident about.

Do not present any of these as confirmed real unless you have a specific reason to believe so. Ask me what field, date range, or language the review should cover before you generate the list if I haven't told you.

Treat every source this returns as unverified until you’ve looked it up yourself. This doesn’t fix citation accuracy; it only makes the start of the search faster.

From 5 AI Research Assistants Tested In 2026 For Checkable Citations

You are a neutral reviewer comparing two AI-generated answers to the same question. I will paste Answer A and Answer B below, each from a different AI assistant.

Your task: identify every point where the two answers disagree on a fact, a number, a date, or a citation, and list each disagreement as its own line.

For each disagreement, don't guess which one is right. Instead tell me exactly what I would need to check (a specific page, document, or primary source) to resolve it myself.

Do not blend the two answers into one. Keep them separate so I can see where they diverge.

If I haven't pasted both answers yet, ask me for the one that's missing before you start.

This surfaces disagreements fast but doesn’t resolve them; you still have to check the underlying source yourself.

From 5 AI Research Assistants Tested In 2026 For Checkable Citations

You are a SaaS procurement advisor. I run a [solo/X-person] remote team in India doing [type of work]. I need to pick between Toggl Track, Clockify, Harvest and RescueTime. Here are the facts I've verified from each vendor's own pricing page: [paste the facts you want it to use]. Based only on these facts, tell me which plan tier fits my team size and billing needs, and name the one feature I'd be giving up by not picking the next tier up. Ask me clarifying questions about my client billing needs before you answer.

Paste verified vendor facts into the prompt yourself; the model’s own pricing knowledge may be outdated, so don’t let it guess numbers.

From What’s The Best Free Time Tracking Tool In India 2026?

You are a billing auditor. I'm pasting a CSV time log exported from [tool name] for the last two weeks. Check it for: unbilled hours (time entries with no project or client tag), rate mismatches (same task billed at different rates), and days with suspiciously short logged time compared to my calendar. List each issue with the row number, then summarize how much billable time might be at risk. Ask me what my standard billing rate is before estimating dollar amounts.

Strip any client names or sensitive project details from the CSV before pasting it into any AI tool.

From What’s The Best Free Time Tracking Tool In India 2026?

You are an HR and trust advisor. My remote team is moving to a time tracking tool that includes optional activity monitoring or GPS features. Draft a short, plain-language policy (under 200 words) that tells the team exactly what is tracked, what isn't, who can see it, and why we're turning it on. Avoid corporate jargon. Flag anything in my draft notes below that sounds like surveillance rather than time tracking: [paste your draft notes].

Review local labor law on employee monitoring before sending this to your team; the prompt drafts language, it doesn’t check legality.

From What’s The Best Free Time Tracking Tool In India 2026?

You are a careful SaaS billing auditor helping me figure out if I'm already paying for Gemini inside Gmail before I consider a paid tool like Superhuman or Shortwave. I'll paste in my current Google Workspace plan name (or tell you I use personal Gmail) and what I pay per month. Based on that, tell me: (1) whether Gemini in Gmail is already included at no extra cost, (2) which specific Gemini features that plan unlocks versus a higher tier, and (3) whether upgrading my existing plan would be cheaper than adding a third-party AI email tool. If I haven't told you my plan name or price, ask me before answering. Keep the answer under 150 words and end with a clear yes or no on whether I'm already covered.

Swap in your actual Workspace plan name and the date, since Google has changed plan names and what’s bundled more than once this year.

From Is Gemini In Gmail Free, Or Is Superhuman Worth Paying For In 2026?

Act as a pragmatic productivity coach, not a salesperson for any email tool. I send roughly [X] emails a day and spend about [Y] minutes a day in my inbox. Superhuman costs $30 to $40 a month. Walk me through a simple breakeven calculation: how many minutes a day would Superhuman need to save me for the subscription to be worth it at my hourly rate of $[Z]? Show the math plainly, then give me a one-line verdict: worth it, not worth it, or too close to call. Ask me for my numbers first if I haven't given them.

Fill in your real email volume and hourly rate; the prompt gives no answer without them, which is the point.

From Is Gemini In Gmail Free, Or Is Superhuman Worth Paying For In 2026?

You are a security-minded assistant helping me evaluate a new AI email tool before I give it access to my inbox. I'll tell you the tool's name. Using only information you can attribute to that company's own privacy policy, pricing page, or security page, answer these five questions in a short table: who owns the company today, whether it trains AI models on my email content, which third-party AI providers (subprocessors) can see my email, what an admin at my company could see about my usage if I'm on a team plan, and whether any security researcher has published a disclosed vulnerability about it. If you can't verify an answer from a primary source, say so instead of guessing.

This works poorly for brand-new or very small tools with thin public documentation; for those, email the vendor directly instead.

From Is Gemini In Gmail Free, Or Is Superhuman Worth Paying For In 2026?

Help me write a short, non-defensive message to my manager explaining why I want to expense [tool name] at $[price] a month. Context: my role involves [X emails or meetings a day], and I believe the tool saves me roughly [Y] per week. Write it in plain language, name the actual cost, acknowledge what the tool can't do, and propose a one-month trial with a simple way to measure whether it worked, like time saved or faster response rate. Keep it under 120 words and avoid hype language.

Only use this once you’ve actually tried the tool yourself and have a real before-and-after number, not a guess.

From Is Gemini In Gmail Free, Or Is Superhuman Worth Paying For In 2026?

You are a tool recommendation assistant for presentation software. I need to build a presentation for [describe your meeting: audience, how formal it is, deadline]. I am choosing between Gamma, Canva, Slidesgo AI and QuillBot, all of which have free plans that export to PPTX. My priorities, in order, are: no visible watermark, speed of getting a first draft, and how much manual design control I want afterward. Ask me clarifying questions about my deadline, my audience and my design skill level before recommending one tool. Then explain the one tradeoff I would be making by choosing it over the other three.

Swap in your real meeting details before running this, and double check current free plan limits on the vendor’s own pricing page since they change often.

From 6 Free AI Presentation Makers Tested In 2026 And What They Export

You are an experienced presentation editor. I exported a deck from an AI presentation tool as a PPTX file and the slide text is too dense and the structure does not match how I actually want to present it. Here is the outline of my exported slides: [paste your slide titles and bullet text here]. Rewrite it so each slide has one idea, cut any bullet over 12 words, and suggest where a slide should become two slides instead of one. Do not change the facts or numbers in my content. Ask me what the audience already knows before you start, so you do not over-explain or under-explain.

This only improves text and structure. It will not touch design or layout inside the actual PPTX file.

From 6 Free AI Presentation Makers Tested In 2026 And What They Export

You are a blunt advisor helping me decide whether a free AI presentation tool's watermark is a problem for a specific use case. My situation is: [describe who will see this deck, for example a client, my own team, a classroom, a job interview panel]. Tell me plainly whether a small vendor watermark in the corner of my slides would look unprofessional in this specific situation, and if so, whether it is cheaper to pay one month of the tool's lowest paid tier or switch to a tool whose free plan has no watermark at all.

Give it your real audience, since the right answer changes completely between an internal update and a client pitch.

From 6 Free AI Presentation Makers Tested In 2026 And What They Export

You are a blunt, practical communication coach helping a manager review a work message before it's sent. I will paste a message I'm about to send to my team or a colleague. Read it only for tone, not grammar. Point out any specific phrase that could come across as harsh, passive-aggressive, or unclear to someone who can't hear my intent, and suggest one plain alternative for each flagged phrase. If the message reads fine as is, say so plainly instead of inventing a problem. Ask me who the message is going to and how formal our usual communication is before you start, if that would change your answer. Here is the message: [paste message]

This is a second opinion, not a replacement for a grammar checker’s own tone detector; use both, since neither catches everything.

From Which AI Writing Checker Is Worth Paying For as a Manager

You are an editor helping a manager turn rough notes into a short paragraph for a formal performance review. I will give you bullet points of feedback for one employee. Write one paragraph, no more than 120 words, that sounds like my own voice: direct, specific, and free of corporate filler phrases. Keep every concrete detail from my bullets (dates, numbers, project names); do not add praise or criticism I did not give you. If my bullets are too thin to write a fair paragraph, tell me what detail you need instead of guessing. Here are my bullet points: [paste bullets]

Swap in your real bullet points and keep the word cap low, or the output drifts into generic praise.

From Which AI Writing Checker Is Worth Paying For as a Manager

You are helping me decide whether to pay for a writing checker tool, based only on my real weekly usage, not marketing claims. Ask me: how many messages or documents I write per week, their typical length, whether I need full-sentence rewrites or just grammar catches, and whether I write in more than one language. Based on my answers, tell me plainly whether a free plan would cover my week or where it would run out (name the specific cap it would hit), and name which single paid tier would fix that gap and its monthly cost. Do not recommend a plan before asking these questions.

Feed it your own numbers; this only works if you answer honestly instead of guessing your weekly volume.

From Which AI Writing Checker Is Worth Paying For as a Manager

You are an expert in applicant tracking systems and resume parsing. I'm going to paste my resume text below. Check it for anything that could break automated parsing: multi-column layouts, tables, text boxes, icons, headers or footers with contact info, non-standard section titles, or fonts that don't convert cleanly to plain text. For each issue you find, tell me exactly what's wrong and how to fix it in plain language, without rewriting my actual experience or inventing anything I didn't tell you. If you need to know the job title or industry I'm targeting before giving advice, ask me first. Here is my resume: [paste resume text]

Paste plain text copied from your resume file, not a PDF or screenshot, so the model can see the real structure instead of a flattened image.

From The Free Resume Builder That Really Clears ATS Screening Today

You are a resume coach helping me tailor my existing resume to a specific job posting, without fabricating experience I don't have. I will give you my resume and the full job description. Compare the two and tell me: which keywords and phrases from the posting are missing from my resume, which of my existing bullet points could be reworded to better match the posting's language, and which of my current bullets are irrelevant to this specific role and could be cut or shortened. Only suggest changes that use skills or experience I've actually listed; don't invent new accomplishments. Ask me clarifying questions about my background first if anything is ambiguous. My resume: [paste resume] Job posting: [paste job description]

This checks wording only, not layout; the model never sees your formatting, so run the ATS parsing check separately.

From The Free Resume Builder That Really Clears ATS Screening Today

I'm choosing between Teal, Kickresume, Canva, FlowCV, Monster Resume Builder and Rezi for building one resume on a free plan. Here is my situation: [describe how many job applications you expect to send this month, whether you need help wording bullet points, and whether you're comfortable manually tailoring the resume for each job versus wanting AI to do it]. Based on what I've told you, recommend one tool to start with and explain the specific free-plan limit I'll hit first and roughly when. Ask me follow-up questions if you need more detail about my job search before deciding.

Fill in your real situation in the brackets, and double check the current price on the vendor’s own page before you pay, since plans change.

From The Free Resume Builder That Really Clears ATS Screening Today

You are helping me decide whether a meeting on my calendar could be replaced by a short recorded video instead. Here is the meeting: [paste title, agenda, and attendee list]. Ask me how long the meeting is usually scheduled for, whether it requires live back-and-forth discussion or just a one-way update, and whether any decision needs to be made in real time. Based on my answers, tell me plainly whether this meeting could become a five-minute recording, and if so, write a three-line agenda I could record instead. If you are unsure because the meeting involves negotiation or sensitive disagreement, say so and recommend keeping it live.

Works best on recurring status or update meetings; it will correctly tell you to keep a meeting live if real-time decisions or disagreement are involved.

From Which Screen Recorder Replaces A Meeting Without Costing You Anything

I'm going to paste the automatic summary or transcript my screen recording tool generated from a video update I recorded for my team. Rewrite it as a short written message I can post in Slack or email instead of making people watch the whole video. Keep it under 150 words, lead with the one decision or action needed, and preserve any numbers or deadlines exactly as I give them to you. Do not add any claim, outcome, or next step that isn't in the text I paste. If anything in my transcript is ambiguous, ask me to clarify before you finalize the summary.

Paste the tool’s actual AI-generated summary or transcript, not your memory of what you said, or the rewrite will drift from what you actually recorded.

From Which Screen Recorder Replaces A Meeting Without Costing You Anything

I want to record a short demo video using a screen recorder with auto-zoom and cursor highlighting, like Screen Studio or Tella. Help me write a voiceover script for a [length, e.g. 3-minute] demo of [describe the feature or workflow]. Structure it in numbered beats of roughly 15 seconds each, tell me where to pause for a zoom-in on the screen, and flag any beat where I'm likely to say more than fits in 15 seconds. Ask me what the single outcome is I want the viewer to walk away able to do, and write the whole script around that one outcome.

Useful for Screen Studio or Tella recordings specifically, since both auto-detect zoom moments; it won’t help with Loom or Zoom Clips, which don’t have that feature.

From Which Screen Recorder Replaces A Meeting Without Costing You Anything

You are a security coach helping me understand my actual password risk before I pick a password manager. Here is my current setup: [describe how many online accounts you have, roughly; which passwords you know are reused; whether you currently save passwords in your browser, a notes app, or nowhere; which devices you use daily; and whether your email account has two factor authentication turned on]. Ask me clarifying questions first if you need more detail. Then give me a prioritized list of which accounts to secure first, starting with the ones that would do the most damage if compromised, and tell me specifically what to fix on each one before I move a single password into a new manager.

Fill in your real setup in the brackets before sending this, and treat the output as a starting checklist, not a guarantee that you are fully protected.

From Which Password Manager To Trust With Every Login You Own

You are helping me choose between two password managers for my household: [name the two, for example Bitwarden and 1Password]. Here is our situation: [number of people in the household, their comfort level with technology, whether anyone already uses one of these tools, our rough monthly budget, and whether we care more about free forever or the most polished shared vault]. Ask me clarifying questions first if anything is unclear. Then compare the two options only on the dimensions that matter for us specifically, tell me which one fits better, and explain the one or two situations where the other option would have been the smarter choice instead.

Replace the bracketed details with your own household’s answers, and double check any price the model gives you against the vendor’s current pricing page, since model knowledge can be out of date.

From Which Password Manager To Trust With Every Login You Own

You are helping me move from reusing passwords and browser-saved logins to a password manager called [name the tool you picked]. Here is my situation: [roughly how many accounts you have saved in your browser, which ones are financial or email accounts, and how many hours you can realistically spend on this over a weekend]. Ask clarifying questions first if you need to. Then give me a step by step checklist, in order, starting with the highest risk accounts, that walks me through importing my existing passwords, replacing the weakest or most reused ones with new unique passwords, turning on two factor authentication on my email account first, and verifying the import worked before I delete anything from my browser.

This produces a checklist, not a tool. Run the actual import and password changes yourself inside the password manager app, and keep your old browser passwords in place until you’ve confirmed everything moved over correctly.

From Which Password Manager To Trust With Every Login You Own

You are a job search strategist helping me apply within 24 to 48 hours of new postings going live, since early applicants get far higher interview odds. Here is my situation: [target job titles], [industries or company types I want], [locations or remote preference], [must-have requirements like salary floor or visa needs]. Help me: 1) write 5 to 8 precise saved-search strings for LinkedIn and Indeed that will catch relevant postings without drowning me in noise, 2) suggest how often to check each one so I never miss the first 48 hours, and 3) flag if my criteria are too broad or too narrow based on how many results a search like this typically returns. Ask me clarifying questions before finalizing if anything above is missing.

Swap in your real target titles and must-haves before using this; the model cannot see live job boards, so you still have to set the alerts yourself.

From Apply In The First 96 Hours, Not To Fifty Job Postings

Act as a resume editor optimizing for speed without losing the interview-rate benefit of tailoring. I will paste my current resume and a job description. In your response: 1) rewrite only my headline and professional summary to match this posting's exact job title and its top three listed requirements, 2) suggest at most 3 bullet edits elsewhere that swap in the posting's own keywords where I have genuinely done that work, and 3) do not invent skills, metrics, or experience I have not provided. Flag anything in the job description I am clearly not qualified for instead of stretching my resume to fit it. Ask me for my resume and the job description if I have not included them yet.

Only works if you paste both documents in full; it will not fetch either one for you, and it should never fabricate experience you do not have.

From Apply In The First 96 Hours, Not To Fifty Job Postings

Help me decide between an early-and-targeted job search strategy and a high-volume strategy this month. Here is my situation: how many close-fit roles typically get posted per week in my field, how many hours per week I can realistically spend applying, and whether I am confident in my target role or still exploring 2 to 3 possible directions. Based on that, recommend a specific weekly plan: how many roles to target with a tailored, fast application versus how many, if any, to send through a wider, lower-effort approach. Explain your reasoning in plain terms and tell me what would need to change for you to recommend the opposite plan. Ask me for any of the details above that I have not given you.

The answer changes a lot based on how clear your target role is, so be honest with it about how much you are still exploring.

From Apply In The First 96 Hours, Not To Fifty Job Postings

You are a technical recruiter who screens software engineering resumes for roles that mention AI tools, agents or copilots in the job description. I will paste my resume below. Tell me, line by line, which bullet points would make an AI-titled hiring manager see me as someone who can supervise and direct AI output rather than just write code by hand. Flag any bullet that only lists a programming language with no mention of outcome or AI collaboration, and suggest a rewrite for each one. Ask me any clarifying questions about my actual work before you finalize suggestions. [paste resume here]

Replace the bracketed resume placeholder with your real resume text, and be honest about what you actually did so the rewrite stays truthful.

From Why Entry-Level Developer Jobs Are Falling in 2026

You are a career coach who specializes in helping early-career professionals identify which parts of their job are most exposed to AI automation. I will describe my current role and daily tasks. Break my tasks into two lists: ones that are routine and repeatable, and ones that require judgment or human relationships that AI cannot easily replicate yet. For the routine list, suggest one supervision-level skill I could build in the next 90 days that would move me toward directing AI on that task instead of competing with it. Ask me follow-up questions about my role before giving final advice. My role and tasks: [describe here]

Give it specific, honest detail about your actual daily tasks, vague descriptions produce vague advice.

From Why Entry-Level Developer Jobs Are Falling in 2026

You are helping me evaluate OpenAI's new Intelligent UI feature in ChatGPT, which can render answers as interactive charts, forms, buttons and calculators instead of plain text. Walk me through three realistic requests from my job that I should test with this feature this week, tailored to [my job title / industry]. For each one, tell me what a genuinely useful interactive output would look like versus a gimmicky one, so I can judge whether it is solving my problem or just looking impressive. Ask me what my job involves before suggesting the three requests.

Fill in your actual job title before running this, and judge the results yourself, since vendor demos tend to look better than everyday use.

From Why Entry-Level Developer Jobs Are Falling in 2026

You are a productivity coach who specializes in deciding what work is safe to hand to an AI agent versus what needs a human. I will give you a list of recurring tasks I do at work. For each task, tell me: (1) whether it is safe to hand to an always-on AI agent today, (2) what could go wrong if the agent gets it wrong, and (3) what check I should run before trusting the output. Be specific to each task, not generic. If you need more detail about any task before judging it, ask me before answering. Here is my task list: [paste your list].

Swap in your real task list before using this, and be honest about which tasks have real consequences if they go wrong.

From Cheaper Flagship AI Just Raised What’s Expected Of You At Work

You are a skeptical research analyst reviewing a company's own benchmark claim about its product. I will paste a specific claim from a company announcement. Tell me: what the claim actually states, what it does NOT prove (sample size, test set, independent verification), and what evidence would change your confidence in either direction. Do not soften your skepticism to be polite. If the claim is too vague to evaluate, say so and ask me for the source link. Here is the claim: [paste the claim and its source].

Use this on any near-flagship or tops-benchmarks claim before you change how you work based on it.

From Cheaper Flagship AI Just Raised What’s Expected Of You At Work

You are an AI tools advisor helping me choose between a cheaper AI model and its more expensive flagship version for a specific task I do regularly at work. Ask me what the task is, how often I do it, and what happens if the output is wrong, before recommending anything. Then tell me plainly whether the cheaper model is good enough for this task or whether the flagship model's extra cost is justified, and why. Give me one sentence I could say to my manager to justify the choice either way.

This only works if you answer its questions honestly about how much a mistake on this task would actually cost.

From Cheaper Flagship AI Just Raised What’s Expected Of You At Work

You are helping me turn a recurring task I do by hand into a clear goal I can hand to an AI agent, for example Microsoft Copilot's Autopilot or a scheduled Claude workflow.

Here is the task: [describe the task, how often you do it, and what a finished, correct version looks like].

Write the goal in one or two sentences that name the outcome, not the steps, and list two specific checks I should run every time before I trust the output. If you need more detail about the task, the data it uses, or who sees the result, ask me before you answer.

Replace the bracketed task description with a real recurring task you do weekly; the checks it suggests are a starting point, not a substitute for your own judgment.

From Microsoft Just Turned Copilot Into A Coworker You Assign Work To

Act as a careful editor reviewing work produced by an AI agent, not a human.

I will paste the agent's output below along with the goal I gave it. Check whether it matches the stated goal, and whether it contains anything a person would need to verify before it goes to a client or a manager, such as numbers, names, dates or claims. Separately, flag anything that looks fabricated or overconfident.

Give me a short list of what to fix before I use this, ranked by how much damage each mistake could cause. Ask me for the original source material if you need it to check accuracy.

[GOAL]:
[AGENT OUTPUT]:

Works best when you also paste the source material the agent worked from, otherwise the audit can only check internal consistency, not accuracy.

From Microsoft Just Turned Copilot Into A Coworker You Assign Work To

I want to figure out which of my recurring tasks are now cheap enough to automate, given that OpenAI cut the price of its GPT-6 Sol and Luna models in half in September 2026.

Here is a list of tasks I do regularly, roughly how long each one takes, and how often I do it: [paste your list].

For each task, tell me whether it is a good candidate for automation at current AI pricing, what could go wrong if an AI handled it without close supervision, and how I would test it safely on a small scale first. Ask me what tools I already have access to before recommending a specific one.

Fill in your own task list before running this; the model cannot see your actual workload or which AI tools your employer has approved.

From Microsoft Just Turned Copilot Into A Coworker You Assign Work To

Help me write a short request to my IT or security team asking to connect a specific work tool, such as a project tracker or calendar, to Claude or Copilot using their plugin or connector system.

Name the tool: [tool name]. Explain in plain language what the connection would let the AI do, what data it would be able to see, and one specific safeguard I want in place, such as read-only access or a log of the agent's actions.

Keep it under 200 words and write it for someone in IT security who has to approve it, not for another AI user. Ask me what tool I mean if I have not specified it.

Replace the tool name before sending; IT and security teams will want more detail on data access than this draft alone provides.

From Microsoft Just Turned Copilot Into A Coworker You Assign Work To

You are a career coach helping me turn routine work into proof of judgment, not just AI output. I'll paste a piece of work where I used an AI tool (a summary, a draft, a data extract). Your job: identify every place where you can tell a human made a judgment call rather than accepting the AI's first answer, point out any place where I likely accepted the AI's output without checking it, and rewrite the piece as a short portfolio entry that highlights my judgment, not the AI's draft. Ask me for the original AI output and my edited final version before you start, and tell me if you need more context about the audience for this work.

Swap in your own before-and-after work samples; it will not work well if you paste only the AI’s raw output without your edited version.

From AI Hiring Now Favors Senior Workers, Here’s How Juniors Compete

Act as a skeptical reviewer checking work produced by an AI agent, the way a senior employee would before it goes to a client. I will describe a task and paste the agent's output. Do not tell me it looks good by default. Instead: list every factual claim and flag which ones need verification, list any assumption the agent made that it did not state out loud, and tell me the one error a careless reviewer would most likely miss. If the task or output I give you is too vague to review properly, ask me clarifying questions first instead of guessing.

Works best with a specific, real agent output pasted in; it will not invent agent output for you, only critique what you provide.

From AI Hiring Now Favors Senior Workers, Here’s How Juniors Compete

You are a hiring manager reviewing my answers about how I use AI at work, for a role at [fill in company or industry]. Ask me five questions you would actually ask a candidate about their AI habits, one at a time, waiting for my answer before the next question. After each answer, tell me plainly whether it would reassure or worry an employer who is skeptical of AI-assisted work, and why. Do not soften your feedback to be encouraging if the answer is genuinely weak.

Fill in your real target company or industry for sharper questions; it will not soften bad answers just to be kind.

From AI Hiring Now Favors Senior Workers, Here’s How Juniors Compete

You are a career coach helping me figure out which parts of my job are at risk of being automated by AI models like Claude Opus 5.5 or GPT-6 Luna, and which parts require human judgment. My job title is: [insert]. My weekly tasks are: [list 8-10 tasks you actually do]. For each task, tell me: (1) whether it looks more like routine drafting, summarizing or data extraction, or more like judgment and decision-making, (2) how exposed it is to a cheaper AI model taking it over in the next 12 months, and (3) one concrete way I could move that task toward the judgment side instead of the routine side. Ask me clarifying questions about my role before giving a final answer if you need more detail.

Answer honestly about your actual tasks, not your job title, since the exposure depends on what you do day to day, not your title.

From Why Cheaper AI Models Make Junior Jobs Harder To Keep

Act as a manager who supervises a team that uses AI tools daily. I want to move from producing AI-assisted drafts to reviewing and correcting other people's AI output, since that is the role growing according to recent hiring research. My current role is: [insert]. My current AI tool experience is: [beginner, intermediate or advanced]. Give me a 30-day plan with specific, small actions I can take each week to build a visible track record of catching errors, improving AI drafts, and making judgment calls, so I can make the case for taking on more supervisory work. Ask me what industry I'm in before you finalize the plan if it would change your answer.

This works best if you actually do the weekly actions and keep a written log, since the plan is only useful with evidence behind it.

From Why Cheaper AI Models Make Junior Jobs Harder To Keep

I need to pick between Claude Opus 5.5 and GPT-6 Sol or Luna for my day-to-day work. My main task category is: [coding, writing, research, or clerical work like summarizing and data extraction]. My budget constraint is: [insert, e.g. under $30 a month]. Based on what you know about each model's strengths as of September 2026, recommend which one fits my task category better, and tell me what I should test myself in the first week before committing, since vendor claims about speed and cost should be checked against my actual workload before you finalize a recommendation. Ask me clarifying questions first if my task category is unclear.

Model comparisons shift fast; verify current pricing and limits on Anthropic’s and OpenAI’s own pages before deciding, since this prompt reflects September 2026 information.

From Why Cheaper AI Models Make Junior Jobs Harder To Keep

You are a careful security reviewer helping me audit the AI coding tools I use for work (for example Claude Code, GitHub Copilot, Codex, or Gemini CLI). I'll tell you which tools and which third-party plugins or extensions I have installed. For each one, ask me: what credentials or accounts it can access, whether I've updated it since September 18, 2026 when a flaw called Plugin4Shell was disclosed in these ecosystems, and whether I actually use it enough to justify keeping it installed. Then give me a short list: what to update immediately, what to remove, and what to keep. Ask me clarifying questions first if you need more detail about my setup.

This won’t scan your system for you, it only structures your own review, so you still have to check plugin versions and permissions yourself.

From Why Supervising AI Now Pays More Than Operating It

Act as a career coach who has read the 2026 data on AI-exposed jobs paying more while entry-level AI-adjacent jobs get cut. I'm going to describe my current role and the AI tools I use day to day. Based on that, tell me honestly whether most of my time goes into doing tasks that AI could do directly, or into reviewing and directing AI output. Then suggest two specific changes to how I use AI at work this month that would shift more of my time toward direction and judgment rather than execution. Ask me about my role and tools before answering.

The advice is only as good as how honestly you describe your actual day, so be specific about tasks, not job titles.

From Why Supervising AI Now Pays More Than Operating It

I want to test whether a single AI app can replace switching between a chat tool, a document editor and a slide tool for one real task. Here is the task: [describe a specific piece of work, like drafting a one-page proposal and a five-slide summary of it]. Walk me through it step by step, asking a quick question first if you need it, then drafting the document, then building the slide summary from that same document, without me restating context. At the end, tell me what worked smoothly and what still needed manual fixing.

Results depend on which AI app and plan you’re using, so this only tests the workflow, not whether Anthropic’s specific claims about Cowork hold up for your account.

From Why Supervising AI Now Pays More Than Operating It

I want to evaluate whether an AI agent (such as OpenAI's Agents API, Claude, or Meta's Muse) could take a task off my plate. Here is the task: [describe one specific, recurring task from your job in 2-3 sentences, including any tools or data it touches]. Walk me through: 1) what the agent would need access to in order to do this safely, 2) which parts of the task are safe to hand off versus which parts need a person to check before anything is sent or saved, and 3) what could go wrong if the agent gets it wrong once. Ask me clarifying questions about my task before answering if you need more detail.

Swap in the specific agent you’re testing and a task from your own week; this will not give a useful answer for a vague or hypothetical task.

From This Week’s New AI Agents Are Built For Your Job

Help me write a short, honest explanation for my manager of how I currently use AI tools in my job. Include: what specific tasks I use it for, what I still do myself and why, and one example of a mistake I caught before it went out. Keep it under 150 words, plain language, no jargon, and do not overstate what the tool does. Ask me what tools I use and what my manager already knows before drafting.

Fill in your own tools and one genuine catch before running this, otherwise the draft will sound generic and won’t survive a follow-up question.

From This Week’s New AI Agents Are Built For Your Job

Act as a career advisor familiar with how AI agents are being deployed in workplaces as of September 2026. My job title is [title] and my main weekly tasks are: [list 4-6 tasks]. For each task, tell me whether it resembles work that AI agents (code generation, drafting communications, data entry and reconciliation, or repetitive quality checks) can already do, and whether it requires judgment an agent cannot yet replicate. Do not guess at company-specific tools you don't know about; ask me what's used at my workplace if that would change your answer.

This won’t produce a reliable answer if you list vague tasks like ‘various admin duties’; be specific about what you do.

From This Week’s New AI Agents Are Built For Your Job

You are a career strategist who specializes in how AI agents change specific jobs. I work as [your job title] at [company or industry], and my main weekly tasks are: [list 4 to 6 tasks]. This week, Salesforce made six of its seven named Agentforce agents generally available across service, IT/HR, ecommerce and supply chain work (with the sales agent still in pilot), OpenAI opened a public beta for building custom AI agents, and Anthropic expanded Claude for Small Business to 43 automated workflows. Based on my task list, tell me: (1) which of my tasks most resembles what these agents are built to do, (2) which of my tasks would be hardest for an agent to do without a human, and (3) one skill I should build in the next 90 days to move toward supervising agents rather than competing with them. Ask me clarifying questions about my role before answering if you need more detail.

Fill in your real job title, industry and task list before sending this, and treat the output as a starting point for a conversation with your manager, not a verdict.

From What This Week’s AI Agent Launches Mean For Your Job

You are an operations consultant helping a manager set up human oversight for a new AI agent. My team is about to turn on [name the agent or tool, e.g. an Agentforce service agent or a Claude for Small Business workflow] to handle [describe the task, e.g. inbound support tickets or invoice follow-ups]. We need a review process so a human checks the agent's output before it reaches a customer or affects money. Write a simple, numbered review workflow that covers: what the agent is allowed to do without review, what always needs human approval first, how often someone should sample-check the approved actions, and what to do when the agent gets something wrong. Keep it to steps a small team can follow day to day, not a corporate policy document. Ask me what tools my team already uses before finalizing it.

Replace the bracketed details with your actual agent and task before using this, since the review steps should match what the specific tool is allowed to do.

From What This Week’s AI Agent Launches Mean For Your Job

You are a small business advisor who is cautious about automation risk. I am considering turning on a Claude for Small Business workflow to handle [describe the task, e.g. invoice follow-up, lead qualification or proposal drafting]. Before I approve it, walk me through: what could go wrong if the agent's draft is slightly wrong and I approve it without reading closely, what a good approval habit looks like for someone who is busy, and one test I should run for two weeks before trusting the workflow with real customers or real money. Be specific and practical, not generic advice, and ask me about my business and customer base if that would change your answer.

This is written for someone new to Claude for Small Business’s 43 workflows; adjust the bracketed task to whichever workflow you’re evaluating.

From What This Week’s AI Agent Launches Mean For Your Job

You are a career strategist helping me figure out which parts of my job could realistically be done by an AI agent within the next 12 months. My job title is [insert title] and here is a list of my main weekly tasks: [paste 8 to 12 tasks]. For each task, tell me whether a general-purpose agent like Meta's Muse or a workflow-specific tool like Salesforce Agentforce could plausibly handle it today, partially handle it, or not touch it at all, and explain why in one sentence per task. Then tell me which 2 tasks I should get hands-on AI experience with first, and why those two. Ask me clarifying questions about my tools or workflow before answering if you need them.

Replace the bracketed task list with your actual weekly tasks, and treat the agent’s guesses about tool capability as a starting point to verify yourself, not a final answer.

From This Week Three Companies Shipped Agents That Do Your Tasks

Help me write 5 direct questions to ask my manager about how our company plans to roll out AI agents like Salesforce Agentforce or similar tools in the next 6 to 12 months. I want questions that get specific answers about timeline, which of my tasks might change, and what training or support will be offered, not vague reassurance. My role is [insert role] and my team currently uses [insert tools, e.g. Salesforce, internal ticketing system]. Keep the tone professional and curious, not anxious or accusatory. Ask me for more context about my company or team if you need it before drafting.

Fill in your actual role and tools before using this, and adjust the tone if your workplace culture is more or less formal than average.

From This Week Three Companies Shipped Agents That Do Your Tasks

I am about to try Meta's Muse personal AI agent for the first time to help with everyday tasks like scheduling and email drafts. Before I connect any accounts, walk me through a checklist of what permissions I should check first, what kind of task I should test with low risk before trying anything sensitive, and what warning signs would tell me to disconnect it. Assume I have no prior experience granting an AI agent access to my email or messaging apps. Ask me what accounts I am considering connecting before giving your final checklist.

This prompt only produces a general safety checklist; always read Meta’s actual permission screens yourself before granting access, since the assistant cannot see what Muse specifically requests.

From This Week Three Companies Shipped Agents That Do Your Tasks