Every day we pick up the prompts and prompting tricks people are sharing, write a reusable version of each one, and run it before it goes on this page.
You get an edited version of your own photo in a new setting, with your face and the number of people left alone. Splitting the brief into subject, setting, lighting and a clear 'keep the same' rule gives the tool less room to drift.
You get a clean six-part breakdown of your old video prompt, plus a single polished version ready to paste into your video tool. It works because each element has its own clause, so nothing is blurred together and gaps are filled openly rather than guessed silently.
It edits a room photo so that one item changes and the rest stays as it was. Naming what must not change before the edit stops the model redesigning the whole room.
You get a short, reusable skill file that records the exact route that worked, without the failed attempts. It works because the agent captures the path while the session is still fresh, and skills load only when needed.
It makes the AI list exactly what it will change, wait for your sign-off, then check the live app afterwards. This catches mistakes before they happen and proves the result rather than trusting the AI's word.
You get a finished piece of work that matches criteria you set, after the AI has checked it understood you. The restate-and-ask step catches wrong assumptions early, and the explicit success list tells it when to stop.
You get a much shorter version of a long log that keeps the errors exactly as written and tells you what was removed. This works because repeated lines are collapsed with counts, while anything important is kept word for word and anything doubtful is flagged for a look at the original.
You get a much shorter version of your code that still keeps validation, security and accessibility. It works because the AI must check what already exists before writing anything new, and must say what it skipped.
You get a short, answer-first reply that keeps the numbers, code and warnings intact. It works because it fixes a strict sentence order and a list of words that must never be cut, so the reply gets shorter without losing meaning.
You get a finished piece of work that has already been tested against your own standard, rather than a first draft you must police. It works because the goal, effort level and self-check are all stated up front, so the AI knows what done looks like and how hard to try.
Each prompt runs on Claude Opus 5.5 with the example shown on its page, and you can read the full reply. A second, strict pass then scores whether the reply did every step and kept to the format. Image prompts are listed, and marked as not yet tested.
We credit and link where each idea came from, and write our own version rather than copying anyone's prompt word for word.