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 Avinava/document-design-system --skill diagram-designgit clone --depth 1 https://github.com/Avinava/document-design-systemWrote 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/avinava/document-design-system/diagram-design)<a href="https://agentmods.dev/skills/avinava/document-design-system/diagram-design"><img src="https://agentmods.dev/badge/skills/avinava/document-design-system/diagram-design/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/avinava/document-design-system/diagram-design"><img src="https://agentmods.dev/badge/skills/avinava/document-design-system/diagram-design.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 | $0.00148 | $0.02517 |
| Opus 5.5 | $0.00059 | $0.01007 |
| Sonnet 5.5 | $0.00030 | $0.00503 |
| Haiku 4.5 | $0.00015 | $0.00252 |
Grade A, and why
diagram-design 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 18d 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 — 157 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Diagram Design
A diagram earns its place when the arrangement itself carries meaning that a sentence or a list cannot carry as fast.
That is the whole test. Boxes-and-arrows restating a numbered list is slower than the list. A flowchart of a linear three-step process is worse than three sentences. Before drawing anything, say what the reader will understand from the shape that they would not get from the prose.
If the answer is "nothing," write the prose.
The standard
The highest-quality move is usually deletion. Every node earns its place. Target density around 4/10 — a diagram that fills its frame is almost always doing too much.
The accent color is reserved for the one or two things the reader should look at first. Everything else is ink and hairlines.
(This standard is adapted from cathrynlavery/diagram-design, MIT. See Credits.)
Choosing a form
Match the form to the relationship being shown, not to the data you happen to have.
| The reader needs to understand | Form |
|---|---|
| How parts are arranged and what talks to what | Architecture / system map |
| How something moves through steps, with branches | Flow |
| Who calls whom, in what order, over time | Sequence |
| What states exist and what transitions between them | State machine |
| How entities relate and what they hold | Data model |
| What happened when, and in what order | Timeline |
| How responsibility is divided across actors | Swimlane |
| How options compare on two independent axes | Quadrant |
| What sits on top of what | Layer stack |
| How something narrows or concentrates | Funnel |
| How sets overlap | Venn |
| How one thing decomposes into parts | Tree / nested |
| What a change adds, removes, and leaves alone | Change view |
| Where each part runs: cluster, namespace, replicas | Deployment map |
| What must finish before what, and which chain sets the date | Dependency graph |
If the answer is "several of these," the diagram is doing too much. Split it. Two clear diagrams beat one complete one — a reader who has to decode a diagram has already lost the time the diagram was meant to save.
What ships with it
12 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.
- core/a11y.md 2.9 KB
- core/base.css 7.5 KB
- core/print.css 4.3 KB
- core/tokens.md 8.6 KB
- references/diagram-families.md 12 KB
- references/importing.md 3.9 KB
- references/mermaid-bridge.md 7.9 KB
- references/primitives.md 12 KB
- scripts/check_diagrams.py 31 KB runs code
- scripts/import_diagram.py 18 KB runs code
- scripts/render_diagram.mjs 7.6 KB runs code
- templates/diagram.svg 4.8 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.
- 18d ago Changed · +8 lines · +9 tokens per session 4613f250a56f
- 1mo ago First seen · 149 lines · 139 tokens per session scan A 78061c92147c
diagram-design is a skill published in the GitHub repository Avinava/document-design-system (2 stars, last pushed 19d ago), licensed MIT. It adds 148 tokens to every session and 2,517 once invoked, about $0.0006 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
ui-ux-audit
Audits web UI and raises it from competent to exceptional. Catches what AI tools ship and people notice at once: broken or shifting layouts, cards and values off a shared line, orphaned words and lone grid items, drifting radii, gutters, ratios and component sizes, text overflow, missing states, poor contrast, small…
ui-ux-laws
Priority-ordered rulebook of core UX/UI laws, cognitive-psychology principles, Nielsen's usability heuristics, and accessibility requirements that Claude must actively apply (not just cite) whenever it designs, builds, redesigns, reviews, or critiques any user interface — websites, web/mobile apps, dashboards, forms…
design-component
Design a UI component spec to the house quality bar (anatomy, variants, sizes, the 8 states, token mapping, and accessibility). Use when the user wants to design or document a component (button, input, tabs, toast, combobox, date picker, modal, etc.) at the spec level before or alongside code. For generating framework…
design-screen
Design or build a page, screen, landing page, website, app or UI from a request - the whole job, end to end. Use this for "design a landing page", "build a settings screen", "make me a website", "design the onboarding", "ออกแบบหน้าเว็บ", "ทำหน้า landing", "ออกแบบ UI", "ทำแอป". It runs the pipeline no single skill…
critique
Adversarial design critique of the current work. Render it, look at it, and argue for rejection. Run after the gates are green, never instead of them.
scaffold-project
Scaffold a new design-product project that matches the recommended Claude Code layout (the reference structure). Use when starting a fresh product/app that will be built with this design system.