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 cosmix/loom --skill loom-accessibilitygit clone --depth 1 https://github.com/cosmix/loomWrote 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/cosmix/loom/loom-accessibility)<a href="https://agentmods.dev/skills/cosmix/loom/loom-accessibility"><img src="https://agentmods.dev/badge/skills/cosmix/loom/loom-accessibility/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/cosmix/loom/loom-accessibility"><img src="https://agentmods.dev/badge/skills/cosmix/loom/loom-accessibility.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Prompt Injection · line 92 Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.Fix: Audit all comments and invisible characters. Remove any instructions that direct the agent to perform unauthorized actions. Use plain, reviewable content.
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 | $0.00016 | $0.04934 |
| Opus 5.5 | $0.00006 | $0.01974 |
| Sonnet 5.5 | $0.00003 | $0.00987 |
| Haiku 4.5 | $0.00002 | $0.00493 |
Grade A, and why
loom-accessibility 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 — 331 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Accessibility
Overview
Making web UIs usable by everyone, including keyboard-only and assistive-technology users. Covers WCAG 2.1/2.2 AA, semantic HTML, ARIA, keyboard/focus management, screen readers, contrast, and testing. Optimize for AA — that is the legal bar (ADA, EN 301 549, AODA) in most jurisdictions.
Core Rules (internalize these first)
- First rule of ARIA: don't use ARIA. A native
<button>/<a>/<input>/<nav>ships focus, keyboard, role, and state for free. ARIA only describes; it never adds behavior. Bad ARIA is worse than none. - Second rule: don't change native semantics.
<button role="heading">is a footgun. Don't override roles of interactive elements. aria-hidden="true"on a focusable element is a trap — it hides the element from the a11y tree while leaving it in the Tab order, so SR users land on "nothing." Hide the whole subtree and remove it from tab order (inert), never one or the other.- Never
outline: nonewithout a replacement. Removing the focus ring with no:focus-visiblestyle is the single most common WCAG 2.4.7 failure. - Automated tools catch ~30–40% of issues. axe/Lighthouse find contrast, missing alt/labels, dup IDs — never "is the focus order sane," "does the SR announcement make sense," "is this keyboard-operable." Manual keyboard + SR passes are mandatory.
Accessible Name Computation (the thing people get wrong)
The accessible name is what a screen reader announces. Resolution order (first non-empty wins):
aria-labelledby(space-separated IDs; concatenated text of targets) — highest priority, overrides visible contentaria-label- Native labeling:
<label for>/wrapping<label>,<img alt>,<fieldset><legend>,<figcaption>, wrapped text content of a button/link title/placeholder— fallback only; never rely on these (placeholder vanishes on input,titleis not exposed on touch/keyboard reliably)
⚠ Gotchas:
aria-labelledbyreferencing a hidden (display:none) element still contributes its text — intentional, useful for SR-only names.aria-label/aria-labelledbyare ignored on non-interactive, non-landmark elements (<div>,<span>,<p>) unless they have a role. Don't expect<span aria-label>to announce.- Icon-only controls MUST have a name:
<button aria-label="Close">✕</button>. A bare glyph/SVG announces nothing or reads the character. aria-describedbyadds supplementary description (hint, error) read after the name — it does not replace the name.
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 Changed · -28 tokens per session 18507fc707dd
- 1mo ago First seen · 331 lines · 44 tokens per session scan A 46734c78f8bb
loom-accessibility is a skill published in the GitHub repository cosmix/loom (62 stars, last pushed yesterday), licensed MIT. It adds 16 tokens to every session and 4,934 once invoked, about $0.0001 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.
Other skills, from other repositories
eval-graphics
Turn an eval study's numbers into on-brand, publish-ready figures using the Newsjack chart room (the eval design system), then validate them with Playwright. For producing the charts in a published eval/data study.
drawio-logical-diagrams
Creates professional logical flow diagrams and logical system architecture diagrams using draw.io XML format (.drawio files). Use when creating: (1) logical flow diagrams showing data/process flow between system components, (2) logical architecture diagrams representing system structure without cloud provider…
aws-drawio-architecture-diagrams
Creates professional AWS architecture diagrams in draw.io XML format (.drawio files) using official AWS Architecture Icons (aws4 library). Use when the user asks for AWS diagrams, VPC layouts, multi-tier architectures, serverless designs, network topology, or draw.io exports involving Lambda, EC2, RDS, or other AWS…
shadcn-ui
Provides complete shadcn/ui component library patterns including installation, configuration, and implementation of accessible React components. Use when setting up shadcn/ui, installing components, building forms with React Hook Form and Zod, customizing themes with Tailwind CSS, or implementing UI patterns like…
tailwind-design-system
Skill for creating and managing a Design System using Tailwind CSS and shadcn/ui. Use when defining design tokens, setting up theming with CSS variables, building a consistent UI component library, initializing a design system configuration, or wrapping shadcn/ui components into design system primitives.
tailwind-css-patterns
Provides comprehensive Tailwind CSS utility-first styling patterns including responsive design, layout utilities, flexbox, grid, spacing, typography, colors, and modern CSS best practices. Use when styling React/Vue/Svelte components, building responsive layouts, implementing design systems, or optimizing CSS workflow.