Not in the top 1,000 by GitHub stars

marcosnovo/unslop-design

An agent skill (Claude Code, Codex, Cursor, Copilot, Gemini CLI...) that makes your app or site stop looking AI-generated: 64 numbered design fixes, scripts that score your UI from 0 to 100 (example: 26 → 100), and a before/after you approve.

Design

History

Review weekly changes to SKILL.md before updating. Dates mark the weeks in which we observed each version; checks apply to the recorded commits.

View weekly changes

`SKILL.md`

Week of Nov 2, 2026 · Line counts unavailable

Before Not checked yet → After Network or broad shell

First observation → cc2bdb80c00d

No earlier version was observed for comparison.

Compare on GitHub

View on GitHub Add to my skills

Verified commit cc2bdb80c00d9e3798b6b6b5ad6e87b716f1856f

14GitHub stars

Not availablestars in 7 days

1skill in this repository

14commits in 30 days

Heat

Insufficient history

Not enough history yet to measure heat. It appears after a few days of data.

Heat never changes the order by stars, and sponsorship never touches it. How heat works

Stars over time

30-day change not yet available: not enough history.

Individual skills in this repository

unslop-design /

How it works

unslop-design

When your agent uses it

Review and redesign an app or website UI to strip out the "slop" (decorative accent color, gradients, glows, noisy badges, emoji as icons, eyeballed spacing, greetings as titles, Title Case, flat hierarchy, repeated numbers) with a system (spacing, type, color roles, elevation) and an orchestrated process: inventory from recordings or screenshots, grayscale pass, parallel product-designer reviewers, a published winning proposal, devs per area, audits and a PR. Covers SwiftUI, Compose and web (dashboards, landing pages, Tailwind), with scripts that scan the code and measure the rendered page. Use it when asked to "review the design", "make it look less AI", "unslop", "polish this screen", "design review", "winning proposal", "audit this dashboard", or when given videos or screenshots of an app or site asking for feedback or a redesign. Also use it to start a design system (spacing, type, palette, shadow scales) or to audit tokens, contrast or consistency across screens.

What it loads into context

Frontmatter

262 tokens

Always available to the agent

SKILL.md

1,792 tokens

When the skill is activated

Reference documents

17,385 tokens

Loaded on request

Token counts are estimates. Scripts run separately; they are not counted as reference context.

The playbook

Principles (non-negotiable)
Any agent (Claude Code, Codex, Cursor, Copilot, Gemini…)
Working agreement (defaults)
Short flow
If the project already has its own design skill

Scripts executed separately

install.sh
scripts/slop-scan.py
scripts/validate.py
scripts/web-audit.mjs

Files

.github/ISSUE_TEMPLATE/config.ymlasset27 bytes
.github/ISSUE_TEMPLATE/feedback.ymlasset537 bytes
.github/ISSUE_TEMPLATE/new-fix.ymlasset807 bytes
.gitignoreasset38 bytes
CHANGELOG.mddoc3,344 bytes
CONTRIBUTING.mddoc1,122 bytes
LICENSEasset1,068 bytes
README.mddoc17,184 bytes
SKILL.mdcore7,340 bytes
agents/openai.yamlasset360 bytes
assets/hero.pngasset912,783 bytes
assets/process.pngasset55,185 bytes
assets/process.svgasset2,153 bytes
assets/refil-mapa.gifasset1,094,315 bytes
assets/refil-perfil.gifasset1,244,618 bytes
assets/social-preview.pngasset234,643 bytes
assets/videos/refil-mapa.mp4asset1,115,116 bytes
assets/videos/refil-perfil.mp4asset1,181,755 bytes
assets/web-before-after.pngasset336,573 bytes
checklist.mddoc18,203 bytes
examples/dashboard-after.htmlasset11,446 bytes
examples/dashboard-before.htmlasset5,026 bytes
install.shruns3,850 bytes
platforms.mddoc3,636 bytes
process.mddoc8,001 bytes
references.mddoc7,029 bytes
scripts/slop-scan.pyruns7,783 bytes
scripts/validate.pyruns1,927 bytes
scripts/web-audit.mjsruns17,489 bytes
spec-card.mddoc1,923 bytes
web.mddoc6,523 bytes

Read SKILL.md


# Unslop design

Three parts: **what to fix** (`checklist.md`, each fix with its reason),
**how to orchestrate it** (`process.md`, rounds with reviewers and devs in
parallel) and **cases and sources** (`references.md`: the before/afters
that shaped this skill and a map of *Refactoring UI*). The spec card every
reviewer fills is in `spec-card.md`. Per-platform notes
(SwiftUI, Compose, web, widgets, watch, car) live in `platforms.md`; websites,
web apps and dashboards have their own playbook in `web.md`. Read the one
you need, not all of them.

Two scripts measure instead of guessing:
- `python3 scripts/slop-scan.py <paths>` — slop patterns in source (web,
  SwiftUI, Compose) with file, line and fix number. Standard library only.

View the complete source on GitHub

Skills included

unslop-design

Repository facts

License: MIT
Created: Nov 3, 2026
Last pushed: Nov 5, 2026
Forks: 0
Open issues: 0
Project activity: Active

Before you install

Network or broad shell

Automated signals, not a guarantee

Checked Nov 6, 2026

No broad tool permissions
4 dangerous shell commandsin 2 files: install.sh:4 README.md:67
37 network referencesin 5 files: README.md:22 install.sh:2 SKILL.md:6 and 2 more
No hidden or obfuscated text
No instruction overrides
4 executable scriptsin 4 files: install.sh:1 scripts/slop-scan.py:1 scripts/validate.py:1 and 1 more
No access to secrets
No requests to vote in Hot

Show all 45 signals by file

`README.md`

Network accessline 22

<p align="center"><sub>Real screenshots of <a href="https://refil.app">Refil</a>, same screen and city, before and after running the skill.</sub></p>

Network accessline 24

<p align="center">   <img alt="Claude Code skill" src="https://img.shields.io/badge/Claude%20Code-skill-111111">

Network accessline 25

<img alt="Claude Code skill" src="https://img.shields.io/badge/Claude%20Code-skill-111111">

Network accessline 26

<img alt="Codex skill" src="https://img.shields.io/badge/Codex-skill-111111">

Network accessline 27

<img alt="Agent Skills standard" src="https://img.shields.io/badge/Agent%20Skills-SKILL.md-111111">

Network accessline 28

<img alt="SwiftUI · Compose · Web" src="https://img.shields.io/badge/SwiftUI%20·%20Compose%20·%20Web-111111">

Network accessline 29

<img alt="MIT license" src="https://img.shields.io/badge/license-MIT-111111">

Dangerous shell operationline 67

curl -fsSL https://raw.githubusercontent.com/marcosnovo/unslop-design/main/install.sh | bash

Network accessline 67

curl -fsSL https://raw.githubusercontent.com/marcosnovo/unslop-design/main/install.sh | bash

Network accessline 94

git clone https://github.com/marcosnovo/unslop-design ~/.claude/skills/unslop-design

Network accessline 105

Kiro: `.kiro/skills/`) and commit it, so your team and cloud sessions get   it too. - **[skills CLI](https://github.com/vercel-labs/skills):**

Network accessline 106

it too. - **[skills CLI](https://github.com/vercel-labs/skills):**   `npx skills add marcosnovo/unslop-design -g`.

Network accessline 107

- **[skills CLI](https://github.com/vercel-labs/skills):**   `npx skills add marcosnovo/unslop-design -g`. - **claude.ai, Claude desktop or ChatGPT:** download `unslop-design.zip`

Network accessline 108

`npx skills add marcosnovo/unslop-design -g`. - **claude.ai, Claude desktop or ChatGPT:** download `unslop-design.zip`   from the [latest release](https://github.com/marcosnovo/unslop-design/releases/latest)

Network accessline 109

- **claude.ai, Claude desktop or ChatGPT:** download `unslop-design.zip`   from the [latest release](https://github.com/marcosnovo/unslop-design/releases/latest)   and upload it as a skill (Claude: Settings → Capabilities → Skills).

Network accessline 110

from the [latest release](https://github.com/marcosnovo/unslop-design/releases/latest)

Network accessline 128

python3 scripts/slop-scan.py src/                      # gradients, glows, emoji, arbitrary values, raw colors… with file:line and fix # node scripts/web-audit.mjs http://localhost:3000       # the rendered page at 390/768/1280 px

Network accessline 129

node scripts/web-audit.mjs http://localhost:3000       # the rendered page at 390/768/1280 px

Network accessline 304

Nothing in the rules: all of them read the same `SKILL.md` (the open [Agent Skills](https://agentskills.io) format); Codex also reads

Network accessline 305

[Agent Skills](https://agentskills.io) format); Codex also reads

Network accessline 348

- *[Refactoring UI](https://www.refactoringui.com/)* by Adam Wathan and

Network accessline 349

Steve Schoger, the source of many of the system rules. - Built while redesigning [Refil](https://refil.app), a fuel price app for   iOS and Android.

Network accessline 350

- Built while redesigning [Refil](https://refil.app), a fuel price app for

`SKILL.md`

Network accessline 6

author: marcosnovo   version: "1.5.0"   homepage: https://github.com/marcosnovo/unslop-design

Network accessline 7

version: "1.5.0"   homepage: https://github.com/marcosnovo/unslop-design ---

Network accessline 8

homepage: https://github.com/marcosnovo/unslop-design

`assets/process.svg`

Network accessline 1

<svg xmlns="http://www.w3.org/2000/svg" width="1200" height="210" viewBox="0 0 1200 210" font-family="-apple-system, BlinkMacSystemFont, 'Segoe UI', Inter, Helvetica, Arial, sans-serif">

`install.sh`

Executable scriptline 1
```
#!/usr/bin/env bash
```

Network accessline 2

# Install unslop-design for every coding agent found on this machine. # #   curl -fsSL https://raw.githubusercontent.com/marcosnovo/unslop-design/main/install.sh | bash

Network accessline 3

# #   curl -fsSL https://raw.githubusercontent.com/marcosnovo/unslop-design/main/install.sh | bash #

Dangerous shell operationline 4

#   curl -fsSL https://raw.githubusercontent.com/marcosnovo/unslop-design/main/install.sh | bash

Network accessline 4

#   curl -fsSL https://raw.githubusercontent.com/marcosnovo/unslop-design/main/install.sh | bash

Network accessline 15

NAME=unslop-design REPO=${UNSLOP_REPO:-https://github.com/marcosnovo/unslop-design}

Network accessline 16

REPO=${UNSLOP_REPO:-https://github.com/marcosnovo/unslop-design}

Dangerous shell operationline 63

if ours "$target"; then echo "  $agent: $target"; run rm -rf "$target"

Dangerous shell operationline 67

if ours "$MAIN"; then echo "  main copy: $MAIN"; run rm -rf "$MAIN"

Network accessline 80

echo "  cloning into $MAIN  (Codex, Cline, Zed)"   run mkdir -p "$(dirname "$MAIN")"   run git clone --quiet --depth 1 "$REPO" "$MAIN"

Network accessline 81

run mkdir -p "$(dirname "$MAIN")"   run git clone --quiet --depth 1 "$REPO" "$MAIN"   [ "$DRY" = 1 ] || touch "$MAIN/$MARK"

Network accessline 82

run git clone --quiet --depth 1 "$REPO" "$MAIN"

`scripts/slop-scan.py`

Executable scriptline 1
```
#!/usr/bin/env python3
```

`scripts/validate.py`

Executable scriptline 1
```
#!/usr/bin/env python3
```

`scripts/web-audit.mjs`

Executable scriptline 1
```
#!/usr/bin/env node
```

`web.md`

Network accessline 41

npm i -D playwright            # once, if the project doesn't have it node scripts/web-audit.mjs http://localhost:3000/dashboard --out unslop-audit

Network accessline 42

node scripts/web-audit.mjs http://localhost:3000/dashboard --out unslop-audit

Network accessline 43

node scripts/web-audit.mjs http://localhost:3000/dashboard --dark --widths 1280

How the signals fit together

Facts from a static read of the files at the pinned commit. They describe what the code does, not why.

Some files could not be read; the facts below still hold.

Signal confidence: 44 facts

Network access in prose of documentationREADME.md:22Low confidence
Network access in prose of documentationREADME.md:24Low confidence
Network access in prose of documentationREADME.md:25Low confidence
Network access in prose of documentationREADME.md:26Low confidence
Network access in prose of documentationREADME.md:27Low confidence
Network access in prose of documentationREADME.md:28Low confidence
Network access in prose of documentationREADME.md:29Low confidence
Dangerous shell operation in code of documentationREADME.md:67Low confidence
Network access in code of documentationREADME.md:67Low confidence
Network access in code of documentationREADME.md:94Low confidence
Network access in prose of documentationREADME.md:105Low confidence
Network access in prose of documentationREADME.md:106Low confidence
Network access in prose of documentationREADME.md:107Low confidence
Network access in prose of documentationREADME.md:108Low confidence
Network access in prose of documentationREADME.md:109Low confidence
Network access in prose of documentationREADME.md:110Low confidence
Network access in code of documentationREADME.md:128Low confidence
Network access in code of documentationREADME.md:129Low confidence
Network access in prose of documentationREADME.md:304Low confidence
Network access in prose of documentationREADME.md:305Low confidence
Network access in prose of documentationREADME.md:348Low confidence
Network access in prose of documentationREADME.md:349Low confidence
Network access in prose of documentationREADME.md:350Low confidence
Network access in the frontmatter of the SKILL.mdSKILL.md:6Low confidence
Network access in the frontmatter of the SKILL.mdSKILL.md:7Low confidence
Network access in the frontmatter of the SKILL.mdSKILL.md:8Low confidence
Executable script in code of a file that runsinstall.sh:1High confidence
Network access in a comment of a file that runsinstall.sh:2Low confidence
Network access in a comment of a file that runsinstall.sh:3Low confidence
Dangerous shell operation in a comment of a file that runsinstall.sh:4Low confidence
Network access in a comment of a file that runsinstall.sh:4Low confidence
Network access in code of a file that runsinstall.sh:15High confidence
Network access in code of a file that runsinstall.sh:16High confidence
Dangerous shell operation in code of a file that runsinstall.sh:63High confidence
Dangerous shell operation in code of a file that runsinstall.sh:67High confidence
Network access in code of a file that runsinstall.sh:80High confidence
Network access in code of a file that runsinstall.sh:81High confidence
Network access in code of a file that runsinstall.sh:82High confidence
Executable script in code of a file that runsscripts/slop-scan.py:1High confidence
Executable script in code of a file that runsscripts/validate.py:1High confidence
Executable script in code of a file that runsscripts/web-audit.mjs:1High confidence
Network access in code of documentationweb.md:41Low confidence
Network access in code of documentationweb.md:42Low confidence
Network access in code of documentationweb.md:43Low confidence

Works with

Agent compatibility, from the files in this repository
Agent	Status	Evidence	Install to
Claude Code	Compatible	Standard SKILL.md SKILL.md	`.claude/skills/<skill-name>/`
Codex	Compatible	Standard SKILL.md SKILL.md	`.agents/skills/<skill-name>/`
Cursor	Compatible	Standard SKILL.md SKILL.md	`.cursor/skills/<skill-name>/`
Gemini CLI	Compatible	Standard SKILL.md SKILL.md	`.gemini/skills/<skill-name>/`
GitHub Copilot	Compatible	Standard SKILL.md SKILL.md	`.github/skills/<skill-name>/`
OpenCode	Compatible	Standard SKILL.md SKILL.md	`.opencode/skills/<skill-name>/`

Native The repository ships a manifest, folder or frontmatter field for this agent.
Compatible A standard SKILL.md that this agent can read.
Unknown No evidence either way.

Derived automatically from paths, manifests and frontmatter. Not tested on each agent.

marcosnovo/unslop-design

History

SKILL.md

Share this page

Heat

Stars over time

Individual skills in this repository

How it works

unslop-design

Frontmatter

SKILL.md

Reference documents

The playbook

Scripts executed separately

Files

Skills included

Repository facts

Before you install

README.md

SKILL.md

assets/process.svg

install.sh

scripts/slop-scan.py

scripts/validate.py

scripts/web-audit.mjs

web.md

How the signals fit together

Works with

You might also use

Leonxlnx/taste-skill

Appllama/appllama-skills

oil-oil/oil-ui

ckryptickunal/OpenDesigner

Kainiko943/beautiful-ui

Dragoon0x/everything-design-taste

`SKILL.md`

`README.md`

`SKILL.md`

`assets/process.svg`

`install.sh`

`scripts/slop-scan.py`

`scripts/validate.py`

`scripts/web-audit.mjs`

`web.md`