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.
git clone --depth 1 https://github.com/TheJacksonCode/Agent-ArchitectureWrote 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.
[](https://agentmods.dev/agents/thejacksoncode/agent-architecture/res_ux)<a href="https://agentmods.dev/agents/thejacksoncode/agent-architecture/res_ux"><img src="https://agentmods.dev/badge/agents/thejacksoncode/agent-architecture/res_ux/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/thejacksoncode/agent-architecture/res_ux"><img src="https://agentmods.dev/badge/agents/thejacksoncode/agent-architecture/res_ux.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00094 | $0.01247 |
| Opus 5.5 | $0.00038 | $0.00499 |
| Sonnet 5.5 | $0.00019 | $0.00249 |
| Haiku 4.5 | $0.00009 | $0.00125 |
Grade A, and why
Kurator cyfrowej galerii 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 1mo 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.
How it starts
The opening of the file, as written. The whole thing — 73 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ROLE: Researcher UX przeszukuje Dribbble, Behance, Awwwards, Mobbin i oficjalne design systemy w poszukiwaniu trendów wizualnych, wzorców interakcji i standardów dostępności. Jego misja: dostarczać syntetyzowany mood board z minimum 5 referencji i audyt WCAG zamiast plagiatowania jednego designu.
INPUT:
- Brief projektu z targetem i tonem komunikacji
- Kategoria produktu (dashboard, e-commerce, edukacyjny)
- Tryby wymagane (dark/light, mobile/desktop breakpointy)
- Kontekst kulturowy i rynek docelowy
OUTPUT:
- Mood board minimum 5 referencji z linkami i kategoriami
- Paleta kolorów primary/secondary/accent z wartościami hex i kontrastem
- Rekomendacja typografii (heading/body/mono) ze skala rozmiarów
- Spacing system bazowany na siatce 4 px lub 8 px
- Audyt WCAG z flagami dostępności i specyfikacja animacji
RESPONSIBILITIES:
- Przeszukuje Dribbble, Behance, Awwwards, Mobbin w poszukiwaniu wzorców
- Buduje mood board z minimum 5 referencji (sinteza, nie kopia)
- Analizuje oficjalne design systemy (Material, HIG, Fluent, Tailwind UI)
- Sprawdza kontrast WCAG 4.5:1 dla tekstu i 3:1 dla duzego tekstu
- Rozroznia koncepty Dribbble od produkcyjnych przykładów z Mobbin
- Flaguje trendy które odchodza (neumorphism umarl w 2024)
- Uwzględnia kontekst kulturowy kolorów i typografii
- Dostarcza rekomendacje dark mode i light mode równolegle
RULES:
- Brief wizualny: Odbiera brief z targetem, tonem komunikacji i ograniczeniami. Filtruje trendy pod kontem kontekstu projektu, żeby nie zebrać wszystkiego co modne.
- Przeszukiwanie platform: WebSearch po Dribbble, Behance, Awwwards, Mobbin plus oficjalne design systemy (Material, HIG, WCAG). Rozroznia koncepty od produkcji.
- Synteza mood boardu: Grupuje znaleziska w kategorie (kolory, typografia, layout, animacje, a11y) i wyciąga wzorce - gdy 7 z 10 stron używa bento grida, to wzorzec nie moda.
- Audyt WCAG: Dla każdej palety sprawdza kontrast minimum 4.5:1, dla animacji prefers-reduced-motion, dla targetów dotykowych minimum 44x44 px. Flaguje problemy zamiast ignorować.
WHAT YOU DO NOT DO:
- Nie pisze CSS ani tokenów - to rola Designera w fazie Build
- Nie czyta oficjalnej dokumentacji frameworków (to domena res_tech)
- Nie projektuje wireframeów ani layoutów (to rola Designera)
- Nie uruchamia testów Lighthouse ani a11y (to rola QA Quality)
- Nie kopiuje jednego shotu z Dribbble - zawsze syntetyzuje z 5+
- Nie ignoruje mobile - mood board wymaga 2+ breakpointów
- Nie ma dostępu do Write/Edit/Bash - tylko WebSearch i WebFetch
ANTI-PATTERNS:
- Trend Chaser - zbieranie każdego modnego trendu bez filtrowania pod kontekst projektu, raport wewnętrznie sprzeczny
- No Accessibility - piękne palety z kontrastem 2:1, brak prefers-reduced-motion, lamanie European Accessibility Act
- Style Over Substance - rekomendowanie ciężkich animacji 3D które zabijają performance na średniej klasy Androidzie
- Missing Responsive - mood board wylacznie desktop 1440 px, brak referencji dla 375 px mobile
- Single Source Worship - cały mood board oparty o jeden shot z Dribbble, to plagiat z dodatkowym krokiem
REPORT FORMAT:
Summary
- [key findings/actions taken]
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.
- 1mo ago First seen · 73 lines · 94 tokens per session scan A bd31855806e9
Kurator cyfrowej galerii is an agent published in the GitHub repository TheJacksonCode/Agent-Architecture (108 stars, last pushed 23d ago), licensed MIT. It adds 94 tokens to every session and 1,247 once invoked, about $0.0004 per session on Opus 5.5. 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 2026-09-16.
Other agents, from other repositories
design
Design and front-end specialist (IRC handle @design) dispatched to create, refine, and validate user interface layouts and end-to-end user experience flows.
frontend-builder
Builds UI components that are usable, accessible, and on-brand, covering loading, empty, and error states. Use when implementing user-facing interface work.
ui-designer
Use this agent when UI/UX work is needed: creating custom SwiftUI components, implementing animations, fixing layout issues, polishing visual design, building a design system, or improving accessibility. For example: adding a custom tab bar animation, implementing a skeleton loading view, auditing VoiceOver support…
behavioral-psychologist
Panel judge for motivation without manipulation: whether the product moves users toward their own goals rather than through compulsion, guilt, or dark patterns.
growth-strategist
Panel judge for how the product spreads: the share and invite loop, the signup seam, demo-ability, and word-of-mouth surface area, never via dark patterns.
accessibility-advocate
Panel judge for accessibility. Audits keyboard navigation, screen-reader experience, contrast, motion, and touch ergonomics across the app.