- Home
- Prompt library
Prompts that are ready to paste
145 prompts across 15 categories, each one written for a real work situation with placeholders for your details and a tip for getting a better result.
Brainstorming and strategy
A chat assistant is a tireless brainstorming partner: it will give you thirty ideas when you are stuck at three, argue the other side of a decision, and play a skeptical customer at 11 p.m. What it will not do is know your market, your numbers, or your customers better than you do. Every prompt here asks it to label what is fact versus assumption so you can tell the difference.
Career and job search
A job search is a writing project with a deadline you do not control. AI is good at the parts that eat evenings: matching your experience to a posting's language, turning rambling descriptions of your work into tight bullets, drafting STAR answers, and rehearsing an interview at midnight. These prompts are built so the AI works from your real experience and tells you where it could not find evidence, instead of inventing it.
Customer communication
Customer messages are where tone matters most and time is shortest. These prompts help you answer an angry email without sounding defensive, deliver a price increase without losing the account, and turn twenty similar tickets into templates your whole team can use. Each one asks for the facts of the situation so the reply is honest and specific, not a form letter.
Data and spreadsheets
Spreadsheets are where AI saves the most time and causes the most quiet damage. It will write a working XLOOKUP in seconds and explain the nested IF you inherited, and it will also miscount, mis-add, and confidently describe a trend that is not in the data. These prompts lean into the first and guard against the second: they ask for test cases, arithmetic shown in full, and a list of what could not be checked.
Education and training
Teaching and training share a problem AI is good at: the same material has to be planned, explained, assessed, and adapted for people who arrive with different knowledge. AI will draft a lesson plan around the objective you give it, write a rubric with observable descriptors, produce three reading levels of the same passage, and turn a dry procedure into a training session with practice built in. What you keep is the judgment: whether the content is correct, whether it fits your learners, and every grade.
Finance and analysis
Finance work is where AI is most helpful and most dangerous in the same breath. It is excellent at explaining, structuring, and drafting: a variance narrative from a table, a budget request that anticipates the CFO's questions, a P&L walkthrough a sales manager will actually read. It is unreliable at arithmetic, it will confidently invent a driver for a variance it cannot explain, and it does not know your chart of accounts. So every prompt here tells the AI to show its math, use only the numbers you gave it, and label anything it inferred.
Hiring and management
Hiring and managing people is where AI helps with the writing and the structure while you keep every judgment. It can turn a rough list into a job description, build a scorecard so every candidate gets the same questions, and draft a performance review from your notes that sticks to observed behavior. What it cannot do is decide who to hire, promote, or let go, and it should never be the reason a decision was made.
Leadership and communication
Leadership communication is one job AI cannot take from you and one it can make a lot easier. It will not know what your team is worried about, which decisions you are willing to own, or how the last reorg went. It will turn your honest notes into a clear all-hands talk track, structure a decision memo so the options are visible, draft the FAQ for a change you are announcing, and argue the other side of your strategy before your skeptics do. The prompts here pull the substance out of you first and draft in plain language, because a message that sounds generated undermines the leader who sends it.
Marketing and content
Marketing is the job where AI output is most visible and most likely to look like everyone else's. It will write a campaign brief, ten headline options, a month of social posts, and a five-email nurture sequence before lunch. The value is in what you feed it: your customers' actual words, your real proof, your point of view. The prompts here are built to pull that out of you first and to make the AI show its assumptions so you can cut the generic parts.
Meetings and summaries
Meetings generate more text than any other part of work: transcripts, notes, chat threads, the document someone sent as a pre-read. These prompts turn that pile into things you can act on: an agenda that ends in a decision, minutes with owners and dates, a decision record that keeps the decision made, a summary written for one specific reader.
Planning and productivity
Planning is where a chat assistant earns its keep fastest. You already know what needs doing; the hard part is turning a pile of half-formed tasks, deadlines, and worries into a sequence you can act on. These prompts do that conversion: paste what is in your head or your inbox, answer a few clarifying questions, and get back a plan with priorities, time estimates, and owners.
Presentations and reports
A good deck or report makes one argument and proves it. Most do not, because they were built by opening the template and filling in slides in order. These prompts start from the decision you want and work backward: the storyline, then slide titles that state conclusions, then speaker notes, then the questions you will get. The report prompts do the same for executive summaries, monthly reports, business cases, and board updates.
Research and learning
AI is a fast research partner and an unreliable one at the same time. It explains anything at the level you ask for, argues both sides of a question, and answers questions about documents you give it. It also invents sources, quotes studies that do not exist, and states stale facts with total confidence. These prompts are built around that split: each one asks the model to separate what it knows from what it is guessing, to cite what it can, and to say 'cannot verify' when it cannot.
Sales and outreach
Sales is where AI earns its keep on the writing and the prep, and where it can get you in trouble if you let it run the relationship. It will draft a cold email in seconds, build a discovery-call plan from three lines of notes, and turn a messy transcript into a clean CRM entry and a same-day recap. What it cannot do is know whether the prospect actually has the problem, so every prompt here asks the AI to separate what it knows from what it is guessing and to leave the claims to you.
Writing and email
Most of the writing you do at work is functional: answer the client, say no to the request, tell the boss what happened, get the thread into a form someone can act on. These prompts handle that kind of writing. Each one tells the AI who to be, what you need, what to leave out, and what to hand back, so you paste, fill in the brackets, and get a draft you can edit in two minutes instead of starting from a blank screen.