AI Prompts Library: Tested Prompts by Profession (2026)

This AI prompts library collects tested prompts by profession. Most prompt lists on the internet were never actually run. Someone asked an AI to generate fifty prompts about a topic, published the output, and moved on. You copy one, paste it, and get something vague back. Some collections are sorted by profession and some by tool, because a prompt written for Google Sheets behaves nothing like a prompt written for a chat window.

It works differently. Every collection linked below was run inside the real tool before it was published.

Use the table below to jump to your role, or read the prompt writing section further down if you want to build your own.

How This AI Prompts Library Was Tested

Every prompt in this library went through the same process before it made the cut:

1. Run in ChatGPT and Claude on the tiers most people actually pay for, so the results reflect real access rather than enterprise features
2. Judged on the quality of the first response, because a prompt that needs heavy correction is not saving anyone time
3. Rewritten with clear variables in square brackets so you can swap in your own details without rewriting the structure
4. Re tested after editing to confirm the output still held up

Where one tool clearly beat the other on a specific task, we say so inside the individual collection rather than pretending the tools are interchangeable.

Jump to Your Collection

Prompts for Accountants and Finance Teams

Finance work is where AI prompts either save you a full afternoon or waste twenty minutes, and the difference is almost always the amount of context you provide. A prompt that says “explain this variance” returns a textbook definition. A prompt that includes the account, the period, the budget figure, and the actual figure returns something you can put in front of a manager without rewriting it.

Our accounting collection is the largest set in this library because finance tasks are repetitive and structured, which is exactly the kind of work language models handle well. It covers month end close, variance commentary, client communication, engagement letters, expense categorisation, tax season checklists, and firm operations. Each prompt uses bracketed variables so you can drop in your own figures.

Read the full set: 100 best ChatGPT prompts for accountants

One warning before you start. Never paste client names, account numbers, or identifiable financial data into a consumer AI tool. Replace real figures with placeholders, get the structure back, then fill in the real numbers inside your own system.

Prompts for HR and Recruiting

HR teams write the same documents repeatedly with small variations, which makes this one of the highest return areas for a prompt library. Job descriptions, interview scorecards, policy summaries, offer communication, and performance review language all follow patterns that a model can reproduce well when the prompt is specific enough.

The prompts in this collection are built to avoid the most common failure in HR output, which is generic corporate language that says nothing. Each one asks for defined structure, a specific tone, and a word limit, because unconstrained output in HR reads like a template and employees notice immediately.

Read the full set: 50 ChatGPT prompts for HR professionals

Prompts for College Students

Students get the most value from AI when they use it to structure their thinking rather than to produce finished work. This collection is built around that principle. It covers breaking down assignment briefs, building essay outlines, turning lecture notes into revision questions, explaining difficult concepts at different levels, and preparing for exams using active recall.

We deliberately left out prompts that write complete essays. Beyond the academic integrity problem, most universities now run AI detection, and the output quality on a full essay is worse than what a student can produce from a solid outline and their own reading.

Read the full set: 80 ChatGPT prompts for college students

Prompts for Google Sheets & Spreadsheet Work

Not every prompt collection belongs to a profession. Some belong to a tool, and spreadsheets are the clearest example. Gemini now runs inside Google Sheets on two separate surfaces: a side panel that builds and edits whole spreadsheets, and an AI function that runs a prompt inside a single cell and fills down a column. Most published prompt lists ignore that split entirely, which is why so many of their prompts return nothing useful. Our Sheets collection is sorted by surface first and job second, covering formulas, data cleanup, categorisation, budgets, variance reporting, and dashboards, with the 2026 quota limits noted where they change what a prompt can realistically do.

Read the full set: 60 Gemini prompts for Google Sheets

How to Write Prompts That Actually Work

Every prompt in this library follows the same five rules. Once you understand them, you can build your own for any task, which is more valuable in the long run than any published list. These principles line up with the official prompt engineering guidance from OpenAI.

1. Give the model a role and an audience

“Write a project update” returns something generic. “You are a project manager writing a weekly update for a client who is not technical” returns the right vocabulary at the right level of detail. The role sets the expertise. The audience sets the complexity.

2. Provide the raw material instead of asking the model to invent it

The single biggest jump in quality comes from pasting your actual notes, figures, or draft into the prompt. A model asked to write from nothing produces plausible filler. A model given real material organises, sharpens, and reformats it, which is usually what you wanted in the first place.

3. Constrain the output format

State how long, in what structure, and in what tone. “In under 200 words, as three bullet points, in a direct professional tone” removes most of the rambling that makes AI output obvious to a reader. Without constraints, models default to long and hedged.

4. Show one example of what good looks like

If you have a previous document you were happy with, paste a section of it and say “match this style”. This single technique closes most of the gap between AI output and your own writing voice, and it works far better than listing style adjectives.

5. Iterate in the same thread instead of starting over

When the first output misses, do not rewrite the whole prompt from scratch. Tell the model what was wrong in one sentence and ask for a revision. The model already holds the context, and correcting is faster than restating.

The Prompt Formula

If you want a single template to fall back on, use this structure:

You are a [role] writing for [audience].
Material to work from: [paste your notes, data, or draft]
Task: [what you want produced]
Length: [word count]
Format: [bullets, table, email, outline]
Tone: [direct, warm, formal, plain English]
Avoid: [jargon, filler, specific words you dislike]

Fill in the brackets and you have a prompt better than most of what gets published online.

Mistakes That Ruin AI Output

Asking for too much in one prompt. A prompt that requests a strategy, a budget, a timeline, and an email in one go will do all four badly. Split it into steps and feed each result into the next.

Accepting the first answer. The first output is a draft. One round of correction usually doubles the quality, and it takes fifteen seconds.

Pasting confidential data. Client names, employee records, financial statements, and student information should never go into a consumer AI tool. Anonymise first, then work.

Using the same prompt across every tool. ChatGPT and Claude respond differently to structure and length. A prompt tuned for one may need adjusting for the other, which is why we note tool differences inside each collection. Anthropic publishes its own prompt engineering documentation for Claude, and the recommended structure differs from OpenAI’s in useful ways.

Not saving what works. When a prompt produces something good, save it with your variables intact. That is how a personal library gets built, and a personal library will always beat a public list.

Frequently Asked Questions

What is an AI prompts library? It is an organised collection of tested instructions you can copy into an AI tool to get a specific result. Instead of writing a prompt from scratch each time, you pick one that matches your task and swap in your own details.

Are these prompts free to use? Yes. Every prompt in this library is free to copy, edit, and use in your own work. Most also work on the free tiers of the major AI tools.

Do these prompts work in Claude and Gemini as well as ChatGPT? Most do. Prompts that rely on file uploads, browsing, or very long documents can behave differently between tools, and where that happens we note it inside the individual collection. For Gemini specifically inside a spreadsheet, the prompts need a different structure entirely, which is covered in our Google Sheets collection.

Which AI tool is best for prompts? It depends on the task. ChatGPT is the most versatile for general work, while Claude tends to hold structure and tone better across long documents. Our comparison articles test these differences directly rather than guessing.

Is it safe to put work information into AI tools? Treat any consumer AI tool as a public place. Remove names, account numbers, and anything identifiable before pasting. Many employers also have their own AI policy, so check it before using these prompts at work.

How often is this AI prompts library updated? New collections are added regularly, and existing ones are re tested when major model updates change the way prompts behave.