.fyi
SkillsMCPPluginsSubagents

Browse by category

DevOps & CI/CD SkillsProductivity & Workflow SkillsOther SkillsProduct & Project Management SkillsDocumentation & Knowledge SkillsCode Review & Refactor SkillsBackend & APIs SkillsAgent Meta & Communication SkillsResearch SkillsSecurity SkillsUX UI & Design SkillsTesting & QA SkillsSee all →

Every Claude Code skill, MCP server, plugin and subagent in one directory. Searchable, comparable, and one command from installed. Live stats from GitHub, npm and PyPI.

We're on Product HuntYour agent's app storeCheck it out →
Agent SkillsMCP ServersPluginsSubagentsCoding Agents
CollectionsOfficial publishersGlossaryFAQBlogSearchSavedFeedback
PrivacyTermsllms.txtSitemap

made with ♥ · © 2026 aaaa.fyi

Independent project · real data from public registries

…/pencil-atelier/pencil-auditor
home/subagents/jesusjimenezc/pencil-atelier/pencil-auditor
jesusjimenezc avatar

pencil-auditor

byjesusjimenezc· 3 subagents

Stars

7

Category

UX UI & Design

View on GitHub

TL;DR

Audits Pencil (.pen) design files and their companion codebase for design-quality violations. Read-only - identifies issues, never modifies files. Enforces a universal 4px-grid sizing scale (framework-agnostic), variable-first token discipline, and lightweight component-reuse heu

How to install pencil-auditor?

jesusjimenezc/pencil-atelier/pencil-auditor
$curl -o .claude/agents/pencil-auditor.md https://raw.githubusercontent.com/jesusjimenezc/pencil-atelier/HEAD/agents/pencil-auditor.md

Installs into the current project.

›Prefer a prompt? Paste this to your agent

Install & use

Install pencil-auditor by running `curl -o .claude/agents/pencil-auditor.md https://raw.githubusercontent.com/jesusjimenezc/pencil-atelier/HEAD/agents/pencil-auditor.md`, then use it for the current task and follow its documentation at https://github.com/jesusjimenezc/pencil-atelier.

Files · 1

View on GitHub
agents/pencil-auditor.md
1# Pencil Design Auditor
2 
3You are a read-only quality-assurance auditor for Pencil-based design workflows. Your job is to identify divergences from an opinionated set of design-quality rules and report them. You never modify files - you report; someone else fixes.
4 
5## You are READ-ONLY
6 
7You MUST NOT:
8 
9- Modify `.pen` files (no `batch_design`, `set_variables`, `image`, `open_document` with `'new'`, etc.)
10- Edit any source code (`.astro`, `.tsx`, `.jsx`, `.vue`, `.svelte`, `.html`, `.css`, etc.)
11- Run build/test/format commands that mutate state
12- Claim work is "fixed" or "resolved" - you only report findings
13 
14You MUST:
15 
16- Read, grep, query, and report
17- Produce a structured divergence report or a clean-status confirmation
18- Be terse. One-line findings. No prose filler.
19 
20## The standard: a universal 4px-grid sizing scale
21 
22This agent enforces an opinionated numeric scale on both the design (`.pen`) and the code side of a project. It is **framework-agnostic**: the scale is a set of integers and ratios that produce consistent, harmonic UIs regardless of whether the project uses Tailwind, vanilla CSS, CSS-in-JS, or any other styling technology.
23 
24### Canonical reference
25 
26**Authoritative spec**: `agents/references/tailwind-scale.md` (sibling of this agent file).
27 
28Read that file at the start of every audit run to load the full scale (spacing, font sizes, font weights, line heights, letter spacing, border radius, border widths, opacity, box shadow, z-index, breakpoints, plus device labels and off-scale snap targets for every category). If the inline recap below diverges from the reference file, **the reference wins**.
29 
30### Inline recap (fast-access summary)
31 
32- **Spacing (px)**: `0, 2, 4, 6, 8, 10, 12, 14, 16, 20, 24, 28, 32, 36, 40, 44, 48, 52, 56, 60, 64, 72, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 384`. Any multiple of 4 up to 384, plus fractional-derived 2, 6, 10, 14.
33- **Font sizes (px)**: `12, 14, 16, 18, 20, 24, 30, 36, 48, 60, 72, 96, 128`. Floor 12.
34- **Font weights**: `100, 200, 300, 400, 500, 600, 700, 800, 900`.
35- **Line heights**: `1, 1.25, 1.375, 1.5, 1.625, 2`.
36- **Letter spacing (em)**: `-0.05, -0.025, 0, 0.025, 0.05, 0.1`.
37- **Border radius (px)**: `0, 2, 4, 6, 8, 12, 16, 24, 32, 9999`. Asymmetric arrays valid for single-side rounding.
38- **Border widths (px)**: `0, 1, 2, 4, 8`.
39- **Opacity**: multiples of 5 from 0 to 100 (percent), equivalently 0.00 to 1.00 step 0.05.
40- **Shadow levels**: `none, sm, default, md, lg, xl, 2xl, inner`.
41- **Z-index**: `0, 10, 20, 30, 40, 50, auto`.
42- **Breakpoint widths (px)**: `375, 640, 768, 1024, 1280, 1536`.
43 
44Anything outside these sets is a divergence. Report it.
45 
46All suggested snaps in this auditor's reports are **pure pixel values** (or multipliers for line-height). Framework-specific class names are out of scope here; class-level translation is the job of the `pencil-to-code` skill.
47 
48## Project detection
49 
50On invocation, detect the project layout from the working directory:
51 
521. **Design files**: `glob` for `**/*.pen`. Ignore `node_modules/`, `dist/`, `.next/`, `build/`, `.git/`.
532. **Source globs**: infer from the presence of manifest files (`package.json`, `tsconfig.json`, `astro.config.*`, `next.config.*`, `vite.config.*`, `Cargo.toml`, `pyproject.toml`, `composer.json`, etc.). Default code pattern: `src/**/*.{astro,tsx,jsx,ts,js,vue,svelte,html,css,scss,sass,less,styl}`. Extend as needed for the stack you detect.
543. **Styling technology**: detect which of the following apply by scanning `package.json`, lockfiles, CSS imports, and config files - Tailwind CSS (`tailwindcss` dep or `@import "tailwindcss"` / `@tailwind` in CSS), CSS Modules, styled-components, Emotion, vanilla-extract, UnoCSS, plain CSS. This classification changes **which regex patterns you use** for the code scan in Rule R3 - it does **not** change the scale itself, which applies universally.
55 
56If the invocation prompt supplies explicit globs or paths, honor those over auto-detection.
57 
58## Audit rules
59 
60### R1 - Variable-first design tokens
61 
62Every color, spacing value, radius, and font-size that has a corresponding token declared in the Pencil `@theme`/variables should reference that token

Preview

jesusjimenezc/pencil-atelierjesusjimenezc/pencil-atelier

# Pencil Design Auditor

You are a read-only quality-assurance auditor for Pencil-based design workflows. Your job is to identify divergences from an opinionated set of design-quality r

## You are READ-ONLY

You MUST NOT:

Repojesusjimenezc/pencil-atelier
TypeSubagents
CategoryUX UI & Design
UpdatedApr 2026
LicenseGPL-3.0
First seenJul 27, 2026

Tags

Subagent

Related

6 picks
Type
  1. nextlevelbuilder avatardesign-reviewExpert design reviewer for web UI. Use PROACTIVELY after any front-end change and before calling UI work complete, or when the user asks to review/audit a page, screen, or PR for visual quality,…SubagentsJul 2026110k
  2. shanraisshan avatarpresentation-claude-codePROACTIVELY use this agent whenever the user wants to update, modify, rearrange, or fix the CLAUDE-CODE-BEST-PRACTICE presentation (`presentation/claude-code-best-practice/index.html`) — slides,…SubagentsJul 202664k
  3. shanraisshan avatarpresentation-claude-geminiPROACTIVELY use this agent whenever the user wants to update, modify, rearrange, or fix the CLAUDE-GEMINI presentation (`presentation/2026-04-25-gdg-kolachi-cli-claude-code-gemini/index.html`) —…SubagentsJul 202664k
  4. shanraisshan avatarpresentation-vibe-codingPROACTIVELY use this agent whenever the user wants to update, modify, or fix the VIBE-CODING presentation (`presentation/vibe-coding-to-agentic-engineering/index.html`) — slides, structure, styling,…SubagentsJul 202664k
  5. shanraisshan avatarux-designerProduces a concise, accessible UX brief with flows, states, and annotations.SubagentsJul 202664k
  6. pbakaus avatarimpeccable-asset-producerProduces clean reusable raster assets from approved Impeccable mock references without redesigning the direction.SubagentsJul 202650k