# Peel and Stick > A sticker-bomb for the back of a laptop lid. List up to eight things you are into, one per line - hobbies, > small habits, jokes about yourself. The page turns each line into a die-cut sticker (shape and colours from > its words, one of 26 little pictures from the everyday words it uses), scatters them across a laptop lid > without overlaps where it can, and lets you drag, retype and re-pick each one. A metered run rewrites every > line as two to five sticker words in a dry, sweet or bold tone and keeps private names and put-downs off the > lid. Built on the SkillSafe skill @eachlabs/sticker-design-generation. https://peel-and-stick.skillsafe.ai/ ## What is free - Plain stickers from your own lines, the lid in five finishes, the seeded scatter, Shuffle, drag-to-move (or arrow keys), hand edits of any sticker's words or picture. - Lines holding an email address, phone number, handle or web address are kept back by the page and never sent. - Three sample lids replay their saved words at no cost. - Downloads: the lid (PNG, SVG), a printable A4 sticker sheet at real size with cut lines (SVG), the words (Markdown, JSON). ## The metered task One task, `stickers`, on gpt-terra, metered per run. The input is string fields: `task`, `things` (numbered lines `N. the thing`), `thing_count` and `tone` (`dry`, `sweet` or `bold`). The reply is one JSON object with `stickers`: one `{n, thing, verdict, text, motif}` per line, in order. `verdict` is `sticker`, `skip-personal` (the line names a private person or holds a personal detail) or `skip-unkind` (it puts someone down). The page checks every reply: one answer per line in order with the line unchanged, two to five words within 24 characters, a word of the line kept, and no numbers, boasts or symbols the line never had. The page picks the picture itself whenever the line's words call one up. ## Links - API tutorial: https://peel-and-stick.skillsafe.ai/api.html - Source skill: https://skillsafe.ai/skill/@eachlabs/sticker-design-generation (eachlabs/skills)