Home › Prompt Finder Quiz
Prompt Finder Quiz
Seven questions and you get three prompt categories matched to your work, plus three ready-to-use prompts pulled straight from the library. About a minute. Nothing is stored and there is no signup.
Sources: Anthropic, prompt engineering overview in the Claude developer documentation, on being clear and direct and supplying the context a new colleague would need (docs.anthropic.com); Sahoo, P. et al., "A Systematic Survey of Prompt Engineering in Large Language Models: Techniques and Applications," which catalogues 41 distinct prompting techniques, arXiv:2402.07927, February 2024 (arxiv.org).
What This Quiz Measures
The quiz sorts you on three things rather than testing knowledge. The first is the output you actually need, meaning whether you are producing documents, short persuasive copy, structured data, code or teaching material. The second is your working context, since a solo freelancer, an in house marketer and a teacher reach for different prompts even when the task looks similar. The third is how much editing you are willing to do afterwards, which determines whether you want a tightly specified prompt or a looser one you will shape by hand.
None of the questions have a wrong answer, and there is no score. The result is a routing decision rather than an assessment.
How Results Map to Categories
Answers pointing at long form documents route to the writing and business categories, where prompts assume structure and length matter. Answers pointing at short persuasive copy route to marketing, email and social media, where the prompts are built around a single message and a defined audience. Structured or analytical answers route to data analysis and finance, where output format carries most of the value.
Specialist answers route to coding, legal or HR, which are narrower and assume domain vocabulary. Teaching answers route to education. Where your answers straddle two families, the result names both rather than forcing a single choice, because most real work sits across categories.
If your answers indicate a regional audience, the result includes the SEA business category. That exists because regional context changes output rather than only tone, and IMDA has noted how quickly generative AI adoption has moved among businesses in the region.
How to Use Your Result
Open the first suggested category and try three prompts rather than reading the whole list. Prompt libraries reward sampling over study, and three attempts tell you quickly whether the category matches the work you actually do. Keep the ones that worked in a personal file with your own edits applied, since a prompt you have adapted once is faster to reuse than one you re-find each time.
If nothing in the suggested categories fits, that usually means the task is more specific than any library prompt. The prompt builder is the next step, assembling a prompt from role, task, context and format rather than starting from a template.
Retake the quiz when your work changes rather than periodically. The result reflects the kind of output you need, so it only shifts when that does, for example on moving from producing content to managing people.
One caveat on the result. It routes you to where the useful prompts are, not to where your work is most valuable. A category that looks obvious for your role may still be the wrong starting point if the writing you actually repeat sits somewhere else, so trust the task you described in the answers over the job title you hold.