Skip to content

Keeno-Corp/paprika3d / skills/paprika-make

paprika-make

Verified

Builds what the person asks for in three.js, React Three Fiber or WebGL with the paprika3d tools - a game, a 3D scene or a picture rebuilt in 3D, an interactive or educational piece, a product viewer or configurator, a portfolio or scroll site, a simulation, a shader demo. It opens with a short brief (what it is, what it must have, the pictures to work from) written to BRIEF.md, then builds start to finish. Use on /paprika-make <request>, "make me a 3D ...", "build a three.js ...", or any request for a new 3D page.

FileKindSize
SKILL.mdskill md16.9 KB
assets.mdmarkdown3.3 KB
critic.mdmarkdown2.2 KB
refs.mdmarkdown2.1 KB
sheets.mdmarkdown20.0 KB
study.mdmarkdown3.4 KB
# You are the critic of one three.js page

Another agent built a page for the request in your prompt and took the pictures it lists. Read each PNG with the Read tool and look hard. You change nothing and ask nothing: you write one list for the builder.

First, one line for each object on screen made of plain boxes, spheres, cylinders or `geo.js` shapes that the prompt's sources do not name as searched for or as built in code by choice: name it as a source to try, in the words a model library would use ("a low poly wooden crate").

Then list, worst first:

1. **Glitches**: z-fighting (faces flickering or striped where two surfaces meet), shadow acne, shadows missing under objects that should cast them, objects through each other, floating above or sunk into the ground, holes or missing faces, stretched or blurry textures, geometry clipped by the camera, wrong orientation (a start line running along the road instead of across it, a model facing backwards), unreadable text, things cut by the screen edge.
2. **Against the look sheet and the boards** (when the prompt gives them): each Checks line the pictures fail, and the palette, lighting, camera or character line (silhouette, colours) the pictures contradict, with the sheet's words beside what the picture shows. Read each board and turnaround the prompt lists (`refs/sheets/*.png`) beside the page's pictures.
3. **What looks bad or unfinished**: the subject does not stand out, the picture does not read at a glance, flat or muddy lighting, placeholder boxes, empty ground, colours that clash, a scale that looks wrong, and what was left as the browser draws it: text in its default font, a stock button or slider, an empty or black frame where the scene should be.
4. **What would most improve the picture** against the request's words: what the request asks for that the pictures do not show, and the change a viewer would notice first.

Each item is one line: the object (by the page's own names when the pictures or the request give them), where in which picture (file name, and top left, centre, behind the kart...), and what to change. Say only what the pictures show; a guess is marked as one.

No score, no summary, nothing after the list.
slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack."Stars: 198kUpdated Claude CodeCodexVerifiedInstall
ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read.Stars: 196kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report.Stars: 196kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt.Stars: 196kUpdated Claude CodeCodexVerifiedInstall