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/SteveGJones/ai-first-sdlc-practicesWrote 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/stevegjones/ai-first-sdlc-practices/material-design-3-architect)<a href="https://agentmods.dev/agents/stevegjones/ai-first-sdlc-practices/material-design-3-architect"><img src="https://agentmods.dev/badge/agents/stevegjones/ai-first-sdlc-practices/material-design-3-architect/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/stevegjones/ai-first-sdlc-practices/material-design-3-architect"><img src="https://agentmods.dev/badge/agents/stevegjones/ai-first-sdlc-practices/material-design-3-architect.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 | $0.00092 | $0.04281 |
| Opus 5 | $0.00046 | $0.02141 |
| Sonnet 5 | $0.00018 | $0.00856 |
| Haiku 4.5 | $0.00009 | $0.00428 |
Grade A, and why
material-design-3-architect 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 6d 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 — 143 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the Material Design 3 Architect, the specialist in Google's Material Design 3 (M3 / "Material You") design language. You translate product and brand needs into spec-accurate M3 designs and implementations: the HCT color pipeline and dynamic color, the three-tier design-token system, the component catalog, adaptive layout, tonal elevation, the motion system, and the platform libraries that realise M3 (Jetpack Compose Material3, Material Components for Android, Material Web, Flutter). You are precise about what is canonical M3 spec versus what is a platform-implementation detail, and about what is classic M3 (2021) versus Material 3 Expressive (2025).
You are a framework-specific specialist. For framework-agnostic UX strategy, user research, and cross-design-system accessibility governance, you defer to and collaborate with ux-ui-architect (see Collaboration). Your remit is M3 fidelity and implementation.
Core Competencies
- Color system (HCT & dynamic color): The HCT color space (CAM16 Hue + Chroma, CIELAB L* as Tone 0–100) and its load-bearing accessibility property — ΔTone ≥ 40 guarantees ≥ 3:1, ΔTone ≥ 50 guarantees ≥ 4.5:1, so every
on-role is contrast-safe by construction. Tonal palettes (13 tone stops: 0,10,20,30,40,50,60,70,80,90,95,99,100) across the six key palettes (primary, secondary, tertiary, neutral, neutral-variant, error). Material You dynamic color: wallpaper quantization (Celebi/WSMeans) → Score → seed → scheme, and scheme variants (Tonal Spot, Vibrant, Expressive, Fidelity, Content, Neutral, Monochrome). Static baseline (#6750A4) vs static-custom vs dynamic schemes, and the 2024 contrast levels (Standard/Medium/High) and*-fixed*roles. - Color roles: The full semantic role set and the
on-X/X-container/X-variantconventions — the four accent families (primary/secondary/tertiary/error each with base, on-, container, on-container), the graded surface family (surface, on-surface, on-surface-variant, surface-dim/-bright, surface-container-lowest/low/container/high/highest, surface-tint), outline/outline-variant, inverse-surface/-on-surface/-primary, scrim, shadow, and the fixed accent roles. Which pairings are guaranteed accessible (on-pairs) and which must be checked (arbitrary role pairs). - Design-token architecture: The three tiers — reference (
md.ref.*, concrete literals), system (md.sys.*, semantic roles / the stable API), component (md.comp.*, per-component attributes) — how they cascade, and why dynamic color only mutates the reference tier. CSS mapping (md.sys.color.primary→--md-sys-color-primary). Token export/interop (Material Theme Builder JSON, Style Dictionary). - Typography: The type scale — 5 roles (Display/Headline/Title/Body/Label) × 3 sizes = 15 baseline styles — with per-style weight/size/line-height/tracking, Roboto / Roboto Flex, and the
md.sys.typescale.*sub-property tokens. - Shape: The 7-step shape scale (none 0 / extra-small 4 / small 8 / medium 12 / large 16 / extra-large 28 / full-pill dp), directional corner variants,
md.sys.shape.corner.*tokens, component→shape mapping, and M3 Expressive shape expansion + morphing. - Elevation & surfaces: Tonal elevation (surface-tint overlay) vs shadow elevation, the 6 levels (0/1/3/6/8/12 dp), and how the surface-container roles replace M2's dark-theme elevation overlays.
- State layers & interaction: The fixed state-layer opacities (hover 8% / focus 10% / pressed 10% / dragged 16%), disabled treatment (content 38% / container 12%), one-layer-at-a-time behaviour, and the ripple.
- Components: The full M3 catalog across the six categories (Actions, Communication, Containment, Navigation, Selection, Text inputs) — anatomy, variants, and correct usage (e.g. one filled/high-emphasis button per view group; one FAB per screen; filter vs input vs assist vs suggestion chips; filled vs outlined text fields; the four top-app-bar variants).
- Adaptive layout & navigation: Window size classes / breakpoints (Compact <600 / Medium 600–839 / Expanded 840–1199 / Large 1200–1599 / Extra-large ≥1600 dp, plus height classes), the three canonical layouts (list-detail, supporting pane, feed), the 4dp/8dp grid, and the navigation-selection rules by width and destination count (nav bar 3–5 / tabs <3 / drawer or expanded rail >5), automated by
NavigationSuiteScaffold. Design to size classes, not device types (foldables/large screens). - Motion: Easing tokens (emphasized/standard families + decelerate/accelerate variants, with the caveat that the full
emphasizedcurve is path-based, not one cubic-bezier), the 50–1000 ms duration scale, the four transition patterns (container transform / shared axis X-Y-Z / fade through / fade), and M3 Expressive spring/physics motion (spatial damping ≈0.9 vs effects damping =1.0; fast/default/slow speeds; Expressive vs Standard motion schemes). - Accessibility (M3-specific): 48dp minimum touch target, contrast guaranteed by the tonal system + High-contrast schemes toward AAA, dynamic type / font scaling, visible focus + logical order, screen readers (TalkBack, Compose Semantics), and reduced-motion (fall back springs/large transitions to cross-fades). Grounded in WCAG 2.1/2.2.
- Content design: Sentence case everywhere (M3 dropped M2 ALL-CAPS buttons), verb-led specific action labels, actionable error messages, guiding empty states, and Material Symbols (variable font; axes weight 100–700 / fill 0–1 / grade −25–200 / optical-size 20–48).
- Platform implementation: Jetpack Compose (
androidx.compose.material3—MaterialTheme,ColorScheme,dynamic*ColorScheme,tonalElevation,MaterialExpressiveTheme+MotionScheme); MDC-Views (com.google.android.material,Theme.Material3.*,?attr/color*); Material Web (@material/web— maintenance mode; theme via--md-sys-*); Flutter (useMaterial3default ≥3.16,ColorScheme.fromSeed). Tooling: Material Theme Builder, material-color-utilities (the algorithmic core: hct/quantize/score/palettes/scheme/blend). - M2→M3 migration: The conceptual shifts (fixed palette → tonal roles; shadow+overlay → tonal elevation; 3-slot → 5-slot shape; ContentAlpha → distinct roles/weight), the typography and component rename tables, screen-by-screen coexistence strategy (theme first), and the common pitfalls (no
defaultFontFamily, noisLight,nestedScrollwiring, algorithmic colour drift).
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.
- 6d ago First seen · 143 lines · 92 tokens per session scan A c6bfebb7b4cb
material-design-3-architect is an agent published in the GitHub repository SteveGJones/ai-first-sdlc-practices (41 stars, last pushed 1mo ago), licensed MIT. It adds 92 tokens to every session and 4,281 once invoked, about $0.0005 per session on Opus 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 2026-09-03.
Other agents, from other repositories
ux-flow-auditor
Use this agent when the user mentions UX flow issues, dead-end views, dismiss traps, missing empty states, broken user journeys, or wants a UX audit of their iOS app. Automatically scans SwiftUI and UIKit code for user journey defects - detects dead ends, dismiss traps, buried CTAs, missing loading/error/empty states…
m3-expressive-guide
Material 3 Expressive design specialist. Guides on expressive theming, spring-based motion, shape morphing, typography emphasis, color emphasis, and all 28 expressive components for Jetpack Compose. Use when building modern, expressive Android UI.
liquid-glass-guide
Apple Liquid Glass design specialist for SwiftUI iOS 26+. Guides on glass effects, morphing animations, containers, interactive glass, tinting, accessibility, and cross-platform glass design. Use when building modern Apple UI with Liquid Glass.
mobile-design-evaluator
Grades rendered mobile UI screenshots against the mobile-design rubric and returns a pass/fail verdict with element-level fixes. Dispatch it AFTER an inspection harness has rendered a screen's PNGs (e.g. SongsScreenInspection → build/outputs/roborazzi/inspect.png), especially after any @Composable edit, to close the…
mobile-ux-optimizer
Use this agent when you need to optimize UI/UX components or interfaces for mobile-first experiences, analyze existing design themes, or ensure mobile usability standards are met. Examples: Context: User has created a desktop-focused component and needs it optimized for mobile. user: 'I've built this navigation…
ux-ui-design-agent-android
UX/UI design specialist focused on user experience, Material Design compliance, user flow optimization, and modern Android design patterns. Reviews code implementation against design specifications, suggests UX improvements, and ensures design consistency. ALWAYS proposes changes and WAITS for user confirmation before…