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 agentmods add skills/humanity4ai/project_human/accessibilitynpx skills add humanity4ai/project_human --skill accessibilitygit clone --depth 1 https://github.com/humanity4ai/project_humanWrote 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/humanity4ai/project_human/accessibility)<a href="https://agentmods.dev/skills/humanity4ai/project_human/accessibility"><img src="https://agentmods.dev/badge/skills/humanity4ai/project_human/accessibility.svg" alt="Measured on agentmods" 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 | $0.00058 | $0.00807 |
| Opus 5 | $0.00029 | $0.00404 |
| Sonnet 5 | $0.00012 | $0.00161 |
| Haiku 4.5 | $0.00006 | $0.00081 |
Grade A, and why
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 3d 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 — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Accessibility Audit
Purpose
Unified WCAG 2.2 accessibility audit covering all 86 success criteria across 4 POUR principles (Perceivable, Operable, Understandable, Robust).
Two modes:
- Crawl mode: Provide
pages: [{url, html}]and get per-page scoring (0-100) with ranking, site aggregate, and detailed per-criterion findings. Manual-only criteria are flagged withmanual_reason. - Session mode: Query the user's required WCAG level (A, AA, or AAA) and receive the complete checklist. The agent enforces all criteria in the checklist throughout the session.
Two analysis engines:
- Regex engine (built-in, zero dependencies): 13 regex-based scoring functions covering the 41 fully automatable criteria (~48% of all 86; per
wcag-criteria.tsautomatable flags). - axe-core engine (optional,
pnpm add axe-core): 57 rule checks providing ~80% automated coverage. Auto-detected at import time — silent fallback if not installed.
Boundaries
Always
- Score real HTML against all automatable criteria
- Return structured per-criterion results with WCAG SC ID references
- Distinguish automated findings from manual-review items
Never
- Claim legal compliance
- Score criteria requiring visual inspection as if automated
- Make external network calls
Level-Aware Scoring
- Color contrast: no Level A requirement; AA=4.5:1 normal text / 3:1 large text (SC 1.4.3); AAA=7:1 / 4.5:1 (SC 1.4.6)
- Level A: 31 criteria (18 automatable, 4 partial)
- Level AA: 55 criteria cumulative (31 automatable, 8 partial)
- Level AAA: 86 criteria cumulative (41 automatable, 15 partial)
References
Implementation guidance and enterprise patterns (merged from wcag-aaa-web-design):
references/wcag-aaa-checklist.md— all 86 WCAG 2.2 criteria with implementation guidance and AAA design implicationsreferences/corporate-design-system.md— AAA-verified color palette with precomputed contrast ratios, type scale, BEM architecturereferences/aria-patterns.md— ARIA widget patterns (modals, tabs, accordions, focus traps)references/form-patterns.md— accessible form design and validationreferences/security-error-handling.md— OWASP-aligned secure error handling (CSP, CSRF, output encoding)references/application-states.md— loading, empty, and error state patternsreferences/data-presentation.md— accessible data table and density patternsreferences/navigation-patterns.md— sidebar, breadcrumb, and wayfinding patternsreferences/responsive-breakpoints.md— mobile-first breakpoint systemreferences/corporate-ux-patterns.md— enterprise UX conventions
What ships with it
28 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.
- CHANGELOG.md 719 B
- LICENSE 1.1 KB
- references/application-states.md 5.9 KB
- references/aria-patterns.md 8.0 KB
- references/corporate-design-system.md 9.7 KB
- references/corporate-ux-patterns.md 4.9 KB
- references/data-presentation.md 3.6 KB
- references/form-patterns.md 6.1 KB
- references/navigation-patterns.md 3.4 KB
- references/responsive-breakpoints.md 5.1 KB
- references/security-error-handling.md 7.5 KB
- references/wcag-aaa-checklist.md 11 KB
- rubric.md 1.3 KB
- scenarios/scenarios.yaml 2.7 KB
- schemas/accessibility.input.json 1.6 KB
- schemas/accessibility.output.json 1.8 KB
- skill.yaml 647 B
- templates/base.html 5.8 KB
- templates/check_contrast.py 9.7 KB runs code
- templates/components.css 14 KB
- templates/data-table.html 5.1 KB
- templates/empty-state.html 2.2 KB
- templates/footer.html 3.3 KB
- templates/header.html 3.7 KB
- templates/main.js 9.9 KB runs code
- templates/responsive.css 4.5 KB
- templates/sidebar-nav.html 5.0 KB
- templates/tokens.css 12 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.
- 3d ago First seen · 59 lines · 58 tokens per session scan A d8d93477d571
accessibility is a skill published in the GitHub repository humanity4ai/project_human (4 stars, last pushed 5d ago), licensed MIT. It adds 58 tokens to every session and 807 once invoked, about $0.0003 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 2026-08-31.
Other skills, from other repositories
accessible-forms
Labels, errors, grouping, and autofill for mobile forms. Use this when building any screen that collects user input — login, signup, checkout, profile editing.
input-validation-a11y
Announcing validation results, shaping error messages, and recovering focus. Use this when implementing client-side validation or async server-side errors in forms.
wcag-2-2-for-mobile
WCAG 2.2 success criteria translated to mobile platforms — gestures, inputs, orientation, focus, and target size. Use this when scoping accessibility acceptance criteria for a mobile feature.
keyboard-and-hardware-input
Keyboard, D-pad, and hardware-switch support on iOS (Full Keyboard Access), Android, Flutter, and React Native. Use this when building screens that must be usable without touch.
switch-and-voice-control
Supporting iOS Switch Control + Voice Control, Android Switch Access + Voice Access, and AssistiveTouch. Use this when building any interactive UI, especially custom gestures, drag-and-drop, or non-standard controls.
rtl-support
Right-to-left (RTL) layout mirroring, bidirectional text, numerals, and accessibility implications on iOS, Android, Flutter, and React Native. Use this when supporting Arabic, Hebrew, Urdu, or Farsi locales.