ChatGPT Prompt Library
A free, searchable library of 200+ ready-to-use prompts for ChatGPT, Claude, and every other AI assistant — spanning writing, marketing, coding, business, education, creative, analysis, and productivity. Every prompt is a proven template with fill-in-the-blank placeholders, a usage tip, and an example. Search, copy with one click, and star your favorites. No signup.
How to Use This Tool
- Browse or search — pick a category tab or type keywords to filter all 200 prompts by topic, role, or use case.
- Open a prompt — click any card to expand the full template, a usage tip, and an example.
- Copy it — hit Copy Prompt to send it straight to your clipboard.
- Fill in the blanks — replace the
[SQUARE_BRACKET]placeholders with your own details, then paste into ChatGPT, Claude, or any assistant. - Favorite & share — star the prompts you reuse to keep them in your Favorites view, and use Share to copy a direct link to one prompt.
- Iterate — refine the result in the chat by changing one detail at a time rather than rewriting the whole prompt.
About the ChatGPT Prompt Library
A prompt library turns prompting from guesswork into a repeatable skill. The same model that produces a generic, forgettable answer for a vague request will produce something genuinely useful when you give it a clear role, the right context, a specific task, and the format you want back. The hard part is remembering how to phrase all of that every time. This library does the remembering for you: 200 prompts, each one a tested template you can copy, fill in, and run in seconds. Instead of staring at an empty chat box, you start from a structure that already works.
The collection spans eight areas of work — writing, marketing, coding, business, education, creative, analysis, and productivity — because the prompting patterns that help in one domain rarely transfer cleanly to another. A blog-outline prompt, a code-review prompt, and a study-plan prompt each need different framing, different constraints, and a different output format. Every template here was written for its specific job: it names the role the model should adopt, asks for the exact deliverable, and tells the model how to present it. The payoff is consistency — you get the same high quality whether you are drafting an email or debugging a function.
Each prompt uses [SQUARE_BRACKET] placeholders for the details only you can supply: your topic, your audience, your data, your goal. Replace them before you run the prompt, and be specific — 'a 600-word post for first-time home buyers' beats 'a blog post.' The usage tip on each card points out the one input that most affects quality, and the example shows a filled-in version so you can see the template in action. The more precisely you brief the model, the less editing you will do afterward.
Great results usually come from iteration, not a single perfect prompt. When an answer is close, change one thing at a time: add a missing piece of context, restate the format, or give a one-line sample of the tone you want. For bigger jobs, chain prompts together — outline first, then expand each section, then polish — so you can review and correct between steps instead of hoping one giant prompt gets everything right. Star the prompts that work for you; a personal shortlist of reliable prompts is the single biggest time-saver for anyone who uses AI daily.
These prompts are free to copy, adapt, and use in your own work, commercial or personal, with no attribution required. If you want to go further — building a custom prompt library tuned to your brand voice, your workflows, and your content pillars — that is exactly the kind of system our AI-Powered Marketing team builds for clients. Pair this library with the AI Prompt Builder to craft a prompt from scratch, the AI Token Counter to size a prompt before you send it, and the LLM Cost Calculator to budget the model behind it all.
Frequently Asked Questions
How do I use this ChatGPT prompt library?
Browse the prompts by category, or type a few words in the search box to filter all 200 by keyword, role, or use case. Click any card to expand the full prompt, a usage tip, and an example. Hit Copy Prompt to send it to your clipboard, then paste it into ChatGPT, Claude, or any other assistant. Replace the [SQUARE_BRACKET] placeholders with your own details before you run it. Star the prompts you like to keep them in your Favorites view, and use Share to copy a direct link to a single prompt.
What makes a great ChatGPT or Claude prompt?
Strong prompts give the model four things: a clear role to adopt, the context it needs, a specific task, and the exact output format you want. Vague requests get vague answers, so name your audience, your goal, and any constraints like length or tone. Concrete examples and a sample of the style you want raise quality more than any clever wording. Every template in this library is built on that structure, which is why each one has bracketed slots for the specifics only you can supply. The more precisely you fill those in, the better the result.
What is the difference between role, task, and context in a prompt?
The role tells the model who to be, such as a senior copywriter or a patient tutor, which shapes its vocabulary and judgment. The context is the background it needs to be useful: your audience, your product, the situation, and any rules. The task is the specific action you want performed, like draft, summarize, critique, or compare. Keeping these three separate makes prompts easier to reuse and edit, because you can swap the context while keeping the role and task fixed. Adding a fourth element, the output format, then locks in how the answer is delivered.
How do I iterate on a prompt that is not working?
Change one thing at a time so you can see what each adjustment does. If the answer is too generic, add more context and a concrete example of what good looks like. If the format is wrong, state it explicitly, such as a table, a numbered list, or a fixed set of labeled sections. If the tone is off, name the tone and give a one-line sample. When a response is close, reply with a short correction instead of rewriting the whole prompt, since the model keeps the earlier context in the conversation.
Can I save my favorite prompts?
Yes. Click the star on any prompt to add it to your Favorites, then use the Favorites tab to see just those. Your favorites and copy counts are stored locally in your browser, so they stay private and persist between visits without any account or signup. Because the data lives on your device, it will not follow you to a different browser or computer, and clearing your browser storage will reset it. To keep a permanent copy, paste the prompts you rely on into your own notes document.
Can I reuse and share these prompts freely?
Yes. These templates are free to copy, adapt, and use in your own work, whether personal or commercial, with no attribution required. Prompts are short functional instructions rather than protected creative works, so you can rewrite them to fit your brand and workflow. Use the Share button to copy a direct link to any single prompt for a teammate. Anything you create with them is yours to keep.
Do these prompts work in both ChatGPT and Claude?
Yes, every prompt here is written to be model-agnostic, so it works in ChatGPT, Claude, Gemini, and most other assistants. The core instructions, the role, task, context, and format, transfer cleanly between models. You may notice each model has its own style, so you might tighten the wording or adjust the requested length to taste. For the strongest results, use the most capable version of whichever model you have and give it the full context the template asks for.
What is prompt chaining and when should I use it?
Prompt chaining means breaking a big job into a sequence of smaller prompts, where the output of one becomes the input to the next. For example, you might first ask for an outline, then expand each section, then polish the tone in a final pass. Chaining beats one giant prompt when a task has distinct stages, when you want to review and correct between steps, or when the full job would overflow the model's attention. It also makes problems easier to debug, because you can see exactly which step went wrong. Many templates in this library are designed to be the first or middle link in such a chain.