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 HarKro753/claude-copy --skill dense-ui-metricsgit clone --depth 1 https://github.com/HarKro753/claude-copyWrote 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/harkro753/claude-copy/dense-ui-metrics)<a href="https://agentmods.dev/skills/harkro753/claude-copy/dense-ui-metrics"><img src="https://agentmods.dev/badge/skills/harkro753/claude-copy/dense-ui-metrics/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/harkro753/claude-copy/dense-ui-metrics"><img src="https://agentmods.dev/badge/skills/harkro753/claude-copy/dense-ui-metrics.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 | $0.00229 | $0.02349 |
| Opus 5.5 | $0.00092 | $0.00940 |
| Sonnet 5.5 | $0.00046 | $0.00470 |
| Haiku 4.5 | $0.00023 | $0.00235 |
Grade A, and why
dense-ui-metrics 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 2mo 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 — 153 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Dense UI Metrics
A single source of truth for the numbers that make productive SaaS UIs feel engineered rather than hand-placed. Every value traces to an industry spec: Inter Dynamic Metrics (Rasmus Andersson) for tracking, IBM Carbon 2x grid + Material Design 3 8dp grid for spacing/density, Bringhurst / Carbon Productive sets for type. The point of encoding it is that "premium" isn't taste — it's a small set of locked relationships. Follow them and the rhythm holds; break one and the eye notices even when the user can't say why.
The model in one breath: a small set of universal laws (this file) applied at a chosen set of anchors (a regime). Laws never change; anchors slide with density. Regimes are named points on a continuous axis, not boxes every app falls into — if a design matches no preset, read its base font size and base padding off the reference and apply the laws at those numbers.
Where things live
- This file — the universal laws (grid, tracking, derived height, degrees of freedom) + the apply-loop + self-check. Read top to bottom; it's the whole method.
references/regimes.md— the regime concept and the full property matrices (Dense + Comfortable): per-role font size, weight, line-height, tracking, padding, height. Read this when you need concrete numbers for a component.references/component-patterns.md— recipes for specific components that don't reduce to a single-line control: multi-line list rows, icon-only nav rails, corner- radius scale, row-separation strategy (zebra vs borders vs whitespace), inline code chips. Read this when building lists, rails, cards, or choosing radii/separators.
How to apply (the loop)
- Pick the regime, then the component role. Regime = density anchor set (Dense /
Comfortable / read-your-own). Then find the role's row in
references/regimes.md— it gives size, weight, line-height, tracking, and padding at once. Don't mix values across roles or across regimes on one screen. - Snap every spacing value to the grid (§Spatial scale). Padding, margin, gap,
offsets. No arbitrary numbers —
13px,18px,60pxare smells. Nearest step wins. - Derive height, don't set it.
H = L + 2·Py(Degree 4). A 36px control is20px line-height + 2×8px, not a magicheight: 36pxwith mystery padding. (Multi- line rows are the exception — seereferences/component-patterns.md.) - Apply tracking by font size (§Tracking). Small text loosens (positive), large
text tightens (negative). Headings left at
0look soft; body over-tightened looks cramped. This is where cheap UIs give themselves away. - Vary only along the degrees of freedom (§Degrees of freedom) when a design needs to breathe or compress. Those axes are the only sanctioned ways to move off the defaults without breaking rhythm.
What ships with it
2 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.
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.
- 2mo ago First seen · 153 lines · 229 tokens per session scan A b1c313ef7248
dense-ui-metrics is a skill published in the GitHub repository HarKro753/claude-copy (18 stars, last pushed 3mo ago), licensed MIT. It adds 229 tokens to every session and 2,349 once invoked, about $0.0009 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 2026-08-30.
dense-ui-metrics compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| dense-ui-metricsHarKro753/claude-copy | Skill | 18 | 3mo ago | MIT | A | 229 | Claude Code |
| top-designwondelai/skills | Skill | 2.6k | 4d ago | MIT | A | 113 | Claude Code, Codex |
| web-typographywondelai/skills | Skill | 2.6k | 4d ago | MIT | A | 128 | Claude Code, Codex |
| scroll-craftnateherkai/scroll-craft | Skill | 4.6k | 1mo ago | MIT | A | 177 | Claude Code |
| designlang-tokensManavarya09/design-extract | Skill | 4.8k | yesterday | MIT | A | 30 | Claude Code |
| refactoring-uiwondelai/skills | Skill | 2.6k | 4d ago | MIT | A | 132 | Claude Code, Codex |
| shiny-bslib-themingposit-dev/skills | Skill | 554 | 4d ago | MIT | A | 87 | Claude Code |