Tell the AI who you are.
Copy-ready prompts for your profession and your goal — every template explains why it works.
Generic prompts get generic answers. The fix is specificity on three axes: who you are (role and context), what you need (goal and output format), and how the AI should work (mechanics like ask-first, fact density, and critique loops). Tellaprompt packages that into copy-ready templates: 70+ prompts across 14 professions and 6 goals, each with placeholders to fill and a one-line explanation of why it works — grounded in 12 documented prompt mechanics.
Start with who you are — or what you want
Why these prompts work when improvised ones don't
Every template on this site is built from the same twelve mechanics — documented practices like role definition (behavior and quality rules, not just "you are an expert"), ask-first (the AI asks clarifying questions before producing), fact density (concrete numbers instead of adjectives) and the brutal critic loop (structured critique before anything ships). Each prompt card names the mechanics it uses, linked to the prompt guide where they're explained in two minutes.
Prompts by profession
How to use a template
- Open your profession (or goal) and pick the template closest to your task.
- Copy it and replace every [PLACEHOLDER] with your specifics — the placeholders are the quality lever, not decoration.
- Paste real context below the prompt: the document, the data, the previous email. Context beats cleverness.
- If the output misses, don't rewrite from scratch — answer the AI's questions, tighten one placeholder, or run the critique prompt on the result.
Frequently asked questions
Do these prompts work with ChatGPT, Claude and Gemini?
Yes. The templates use plain-language mechanics (role, context, format, examples, critique) that all current assistants follow — nothing vendor-specific. Where a placeholder matters more on one platform, the prompt's note says so.
Why do the prompts have [PLACEHOLDERS] instead of being ready to send?
Because the specifics are what make output good. A template with your role context, your audience and your constraints filled in reliably beats a longer prompt written for nobody in particular. Filling 3–5 placeholders takes under a minute.
What are the "mechanics" each prompt names?
Twelve documented practices that make prompts work — role definition, ask-first, output format, fact density, precedent examples, the Osborn variation levers, the brutal-critic loop and more. Each is explained with a copy-ready snippet in the prompt guide.
Is this site free?
Yes — all prompts and guides are free to use and free to cite with attribution, in all six languages.
My profession isn't listed — what now?
Start from the goal pages (report, summary, email, campaign, meeting, analysis) — they're profession-agnostic. The roles list grows; the prompt guide teaches you to adapt any template to your field in a few minutes.