Наталя ХандусенкоAI Eng
15 October 2026, 12:30
2026-10-15
A Ukrainian IT professional developed Unslop — an auditor plugin that cleans AI design from patterns and bad decisions
Product and UX/UI Designer Oleg Kuharuk created Unslop — an auditor plugin that will help solve a popular problem in modern UI design and rid AI-generated interfaces of typical errors, outdated patterns, and template solutions (the so-called design slop).
Product and UX/UI Designer Oleg Kuharuk created Unslop — an auditor plugin that will help solve a popular problem in modern UI design and rid AI-generated interfaces of typical errors, outdated patterns, and template solutions (the so-called design slop).
The tool is designed to integrate with AI agents, including Claude Code, Codex, Cursor, and Lovable.
How does Unslop work?
Unlike conventional generators, Unslop functions precisely as an inspector and quality controller:
the tool analyzes a design or layout that the AI agent has already generated;
Instead of a vague assessment based on "eyeballs" or subjective taste, Unslop identifies specific problems and provides the result in file:string format along with a clear fix plan.
The auditor itself does not automatically change the code — the final decision on making edits and the design concept remains with the human.
Dynamic criteria system through MCP
The main technical feature of Unslop is the lack of a hard-wired local rule list.
The plugin does not contain a local copy of the "slop criteria" directory, but directly accesses Oleg Kuharuk's MCP. Each run is recorded by version and hash sum, which eliminates the hidden use of outdated local files.
This means that once a rule is updated or added on the MCP server, it instantly becomes available to all AI agents the next time an audit is run, without the need to reinstall the plugin.
49 criteria for "quality design"
Currently, Unslop’s knowledge base consists of 49 criteria against which each interface screen is tested. All of them are divided into three key categories.
Texts and copywriting: check for the absence of fabricated social proof, artificial creation of haste (false urgency), office notes, or stubs in the production version.
Interface: visual hierarchy control, absence of excess grids and tiles for each icon, checking text contrast, correct concentric radii, optical centering, animations, and sizes of closed touch targets.
Processes and consistency: checking compliance with the design system, preserving working contexts, and cleanliness of component implementation.
A Ukrainian designer has developed the KidRush app — a visual timer for children that will help avoid stress during school and kindergarten preparations
A Ukrainian designer has launched a news application in which AI compiles digests of the most important news from Ukrainian and world media every night. How NAIA works
Як нейромережі бачать вільну та незалежну Україну? Тест dev.ua
Нейронні мережі для генерації зображень бачать світ по-своєму, їхню логіку зрозуміти часом зовсім неможливо. Але таки хочеться. На честь Дня Незалежності України редакція dev.ua вирішила провести невеликий експеримент.
Ми задали чотирьом різним нейронним мережам п’ять однакових запитів: «прапор України», «День Незалежності України», «український Крим», «перемога України» та «українці». Отриманими результатами ми ділимося з вами нижче.
У TikTok тепер можна генерувати фон за допомогою нейромережі. Ми протестували її та ділимося результатами
У TikTok з’явилася нова функція «Розумний фон». З її допомогою як фон для тіктоків можна підставляти згенеровані нейромережею зображення. Редакція dev.ua протестувала цю технологію і ділиться своїми враженнями.