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.
npx skills add abhixv/m3-expressive-design-skill --skill m3-expressivegit clone --depth 1 https://github.com/abhixv/m3-expressive-design-skillWrote 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/skills/abhixv/m3-expressive-design-skill/m3-expressive)<a href="https://agentmods.dev/skills/abhixv/m3-expressive-design-skill/m3-expressive"><img src="https://agentmods.dev/badge/skills/abhixv/m3-expressive-design-skill/m3-expressive/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/skills/abhixv/m3-expressive-design-skill/m3-expressive"><img src="https://agentmods.dev/badge/skills/abhixv/m3-expressive-design-skill/m3-expressive.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 | $0.00077 | $0.02151 |
| Opus 5.5 | $0.00031 | $0.00860 |
| Sonnet 5.5 | $0.00015 | $0.00430 |
| Haiku 4.5 | $0.00008 | $0.00215 |
Grade A, and why
m3-expressive 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 — 128 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Material 3 Expressive
M3 Expressive is an evolution of Material 3, not a new version — M3 is not deprecated and this isn't "M4". It adds features, updated components, and design tactics for emotionally impactful UX. The Usability guidelines name its design tactics as containment, size, shape, color, and typography — a non-exhaustive list; the blog's seven tactics also cover motion and component flexibility.
Backed by Material's most-researched update since 2014 (46 studies, 18,000+ participants). The headline finding is that this is a usability mechanism, not a decorative one: participants spotted key UI elements up to 4× faster in expressive screens. So every expressive choice must answer one question — does this guide attention to what matters on this screen? If not, it's noise.
Source contract — read this before writing any spec
M3 has exact published values. Stating a plausible-sounding number instead of the real one is the main way this work goes wrong, and it is invisible to the reader.
Every number, token name, and component name you state comes from this skill's references. Look it
up. tokens.md has the numeric bedrock; components/*.md has per-component placement and behavior;
component-tokens.md has the per-component dp geometry — including the measurements m3.material.io
publishes only inside images, so check there before concluding a value is unavailable.
When a value you need is genuinely not in M3, write it as two labelled parts:
From M3: button label uses
label large(14sp / 20sp / 0.1sp tracking, Medium). My design decision (not in M3): 12dp gap between the two action buttons.
That split is the deliverable's shape. It costs one clause and it is the difference between a spec someone can build from and a spec that quietly invents a design system.
M3 publishes no stagger values, no per-component choreography timings, and no reduce-motion scheme-swap rule. Those are always your design decisions.
What ships with it
60 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- README.md 1.3 KB
- references/color-schemes.md 52 KB
- references/color.md 9.0 KB
- references/component-tokens.md 46 KB
- references/components/bars.md 33 KB
- references/components/buttons.md 45 KB
- references/components/containment.md 70 KB
- references/components/fabs.md 29 KB
- references/components/feedback.md 64 KB
- references/components/input.md 61 KB
- references/components/navigation.md 48 KB
- references/expressive-tactics.md 6.5 KB
- references/foundations.md 53 KB
- references/interaction.md 45 KB
- references/layout.md 59 KB
- references/motion.md 4.0 KB
- references/platforms.md 4.4 KB
- references/shape.md 5.7 KB
- references/tokens.md 9.1 KB
- references/transitions.md 26 KB
- references/typography.md 8.7 KB
- references/visuals.md 3.8 KB
- scripts/refresh-component-tokens.js 8.7 KB runs code
- scripts/refresh-m3.js 10 KB runs code
- visuals/building-with-m3-expressive/break-from-the-surrounding-shape-style-to-draw-attention-to-a-particul.png 1140 KB
- visuals/building-with-m3-expressive/expressive-app-bars-toolbars-and-buttons.png 402 KB
- visuals/building-with-m3-expressive/expressive-typography-can-enhance-hierarchy-and-add-personality.png 9503 KB
- visuals/building-with-m3-expressive/group-similar-content-into-informative-groupings.png 783 KB
- visuals/building-with-m3-expressive/material-design.jpg 23 KB
- visuals/building-with-m3-expressive/smaller-shapes-can-result-in-essential-actions-looking-less-important.png 1148 KB
- visuals/building-with-m3-expressive/the-expressive-color-system-can-create-a-wide-range-of-accessible-pale.png 3043 KB
- visuals/building-with-m3-expressive/the-shape-library-has-35-iconic-shapes-which-can-be-used-to-add-visual.png 446 KB
- visuals/building-with-m3-expressive/the-updated-type-scale-includes-new-emphasized-text-styles.png 551 KB
- visuals/building-with-m3-expressive/ungrouped-information-can-blend-together.png 768 KB
- visuals/building-with-m3-expressive/use-contrast-to-emphasize-the-main-takeaway-or-element.png 1350 KB
- visuals/building-with-m3-expressive/without-contrast-elements-can-blend-together.png 1347 KB
- visuals/components_all-buttons/a-button-s-level-of-emphasis-helps-determine-its-appearance-typography.png 24 KB
- visuals/components_all-buttons/a-filled-button-for-a-high-emphasis-actiona-text-button-for-a-low-emph.png 306 KB
- visuals/components_all-buttons/avoid-placing-a-button-below-another-button-if-there-s-space-to-place-.png 53 KB
- visuals/components_all-buttons/buttontoggle-buttonicon-buttontoggle-icon-buttonsplit-buttonstandard-b.png 106 KB
- visuals/components_all-buttons/for-multiple-actions-choose-a-higher-emphasis-button-for-the-more-impo.png 74 KB
- visuals/components_all-buttons/use-a-filled-button-on-its-own-for-a-single-important-action.png 27 KB
- visuals/components_all-buttons/when-using-multiple-buttons-you-can-place-a-text-button-low-emphasis-n.png 74 KB
- visuals/components_all-buttons/when-using-multiple-buttons-you-can-place-an-outlined-button-medium-em.png 76 KB
- visuals/components_app-bars/a-leading-element-and-a-trailing-element-outside-searcha-leading-eleme.png 39 KB
- visuals/components_app-bars/adjust-the-width-of-the-app-bar-orbiter-to-fit-in-a-person-s-field-of-.png 3331 KB
- visuals/components_app-bars/always-align-the-app-bar-orbiter-within-the-bounds-of-nearby-spatial-p.png 370 KB
- visuals/components_app-bars/an-app-bar-s-accessibility-label-can-incorporate-its-ui-text-as-well-a.png 444 KB
- visuals/components_app-bars/an-icon-button-should-be-clearly-labeled-on-the-action-it-takes-like-v.png 442 KB
- visuals/components_app-bars/app-bar-color-roles-used-for-light-and-dark-themes-surfaceon-surfaceon.png 114 KB
- visuals/components_app-bars/app-bars-can-display-one-high-visibility-action-to-boost-its-prominenc.png 214 KB
- visuals/components_app-bars/app-bars-provide-content-and-actions-related-to-the-current-page.png 157 KB
- visuals/components_app-bars/app-bars-show-information-about-the-page-key-actions-and-navigation-ac.png 249 KB
- visuals/components_app-bars/avoid-expanding-the-app-bar-orbiter-beyond-the-adjacent-panel-s-width-.png 3333 KB
- visuals/components_app-bars/avoid-using-custom-color-roles-for-the-search-bar-container-and-search.png 11 KB
- visuals/components_app-bars/baseline-variantsmediumlarge.png 25 KB
- visuals/components_app-bars/center-aligned-app-barleft-aligned-app-bar.png 56 KB
- visuals/components_app-bars/containerheadlinetrailing-iconsleading-icon.png 76 KB
- visuals/components_app-bars/containerheadlinetrailing-iconssubtitleleading-button.png 20 KB
- visuals/components_app-bars/containerleading-buttontrailing-elementsheadlinesubtitle.png 67 KB
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 · 128 lines · 77 tokens per session scan A 82bf71dff32f
m3-expressive is a skill published in the GitHub repository abhixv/m3-expressive-design-skill (25 stars, last pushed 17d ago), licensed MIT. It adds 77 tokens to every session and 2,151 once invoked, about $0.0003 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 2026-08-31.
Other skills, from other repositories
cinematic-scroll
Design and build cinematic websites, 3D websites, interactive portfolios and product landing pages with scroll-driven storytelling, parallax, text animation and optional Three.js/WebGL scenes. Use for new experiences, motion improvements, scroll audits and storyboards in standalone HTML or an existing app. Includes…
cinematic-scroll
Design and build cinematic websites, 3D websites, interactive portfolios and product landing pages with scroll-driven storytelling, parallax, text animation and optional Three.js/WebGL scenes. Use for new experiences, motion improvements, scroll audits and storyboards in standalone HTML or an existing app. Includes…
html-effectiveness-proto
A guide for testing motion and short interaction flows in a web interface. It covers adjustable animation timing and easing, which controls how movement speeds up or slows down, plus clickable flows across four to six screens.
ai-image-frontend-ui
A workflow for creating frontend user-interface visuals with AI, from visual reference and real materials to generated code, browser screenshots, and visual comparison. It focuses on layouts, typography, exact text, interface states, and avoiding invented content.
huashu-design
A design workflow for making high-fidelity HTML prototypes, slide decks, animations, visualizations, and design reviews. It requires three initial design directions before developing the selected one.
proto-bot-getting-started
A first-conversation guide for starting an interface-prototyping task. It gathers the application, design rules, allowed changes, and preferred delivery format only as needed before producing an initial result.