roll-lint

roll-lint is a plugin for Claude Code from niharya/skills-drawer. Its manifest loads nothing; the 1 skill it bundles cost 201 tokens per session together, scanned A, original, MIT.

A tool that reviews CSS and SCSS stylesheets against design-system patterns. It checks areas such as spacing, corners, typography, layout, accessibility, and naming, then records the patterns left after cleanup.

In plain words
What is it for?
Use it to audit frontend styles, remove design-pattern inconsistencies, and document the tokens and rules that remain.
Why use it?
It helps reveal inconsistent styles and organize cleanup work across a stylesheet.

Plugin for Claude Code

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Claude Code
/plugin marketplace add niharya/skills-drawer
agentmods
npx agentmods add plugins/niharya/skills-drawer/roll-lint
Clone the repo
git clone --depth 1 https://github.com/niharya/skills-drawer

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for roll-lint

README.md
[![agentmods](https://agentmods.dev/badge/plugins/niharya/skills-drawer/roll-lint.svg)](https://agentmods.dev/plugins/niharya/skills-drawer/roll-lint)
Your own site
<a href="https://agentmods.dev/plugins/niharya/skills-drawer/roll-lint"><img src="https://agentmods.dev/badge/plugins/niharya/skills-drawer/roll-lint.svg" alt="Measured on agentmods" height="20"></a>
Per session not measured What this adds to a session before it is invoked.
When invoked not measured The manifest loads nothing itself; its 1 skill cost 201 tokens a session between them.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Security

Grade A, and why

roll-lint scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

roll-lint/.claude-plugin/plugin.json · 22 lines

What it actually says

{
  "name": "roll-lint",
  "version": "1.0.0",
  "description": "Systems-design-aware CSS/SCSS auditor. Cleans stylesheets through seven dependency-ordered phases — spacing, radius, type, layout, accessibility, naming — and returns a blueprint of the patterns that survived the cleanup.",
  "author": {
    "name": "Nihar Bhagat",
    "email": "[email protected]"
  },
  "license": "MIT",
  "repository": "https://github.com/niharya/skills-drawer",
  "keywords": [
    "css",
    "scss",
    "linter",
    "design-system",
    "audit",
    "frontend",
    "design-tokens",
    "stylesheets"
  ]
}
Contents

What it installs

The manifest is a name and a version. 1 skill travel with it, and installing the plugin installs all of them — 201 tokens a session between them. Each is measured on its own page, and each can be installed alone.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 3d ago First seen · 22 lines scan A 7d9bbd074038

Subscribe to this mod's changes

roll-lint is a plugin published in the GitHub repository niharya/skills-drawer (2 stars, last pushed 1mo ago), licensed MIT. Its token cost is not measured: this kind of file is read by the harness, not the model. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other plugins, from other repositories

superdesign-skill

Design or redesign frontend UI, presentations, and marketing graphics on the Superdesign infinite canvas. Uses approved slide outlines, project context, and branchable drafts you refine.

superdesigndev/superdesign-skill · not measured

google-fonts

Google Fonts typography system generator. Searches 1,923 fonts, suggests singles or pairs, generates CSS/Tailwind typographic systems.

sliday/google-fonts-skill · not measured

cinematic-scroll

Build cinematic, scroll-driven websites — pinned chapter reveals, multi-depth parallax, 3D tilt, environment-morphing backgrounds, and full release/launch pages. The motion grammar is the skill; the aesthetic is always the user's. Single self-contained scroll sections (Mode A) or a full Next.js release site with…

MustBeSimo/cinematic-scroll-skill · not measured

uisandbox

Play with your app's design, right in the browser: open a built web app in UISandbox, 1:1 and with its own knobs, from the chat; try a change, verify, export the patch. Bundles the /uisandbox skill and the uisandbox MCP server.

Ideelab/uisandbox · not measured

make-my-website-non-ai-looking

Two skills that build websites which look designed rather than generated, and fix the ones that already look generated. Thirty giveaway patterns tiered into bans, a budget, and requirements.

whatsuppiyush/make-my-website-non-ai-looking-claude-skill · not measured

design-from-code

Workflow skill that designs UI changes from real source code and data models — verify data in code, iterate faithful HTML mockups, lock a design doc, then delegate implementation. It reproduces existing components pixel-for-pixel by reading actual JSX/queries instead of imagining them.

nlook-service/design-from-code · not measured