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 jhlee0409/omni-harness-kit --skill accessibility-checksgit clone --depth 1 https://github.com/jhlee0409/omni-harness-kitWrote 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/jhlee0409/omni-harness-kit/accessibility-checks)<a href="https://agentmods.dev/skills/jhlee0409/omni-harness-kit/accessibility-checks"><img src="https://agentmods.dev/badge/skills/jhlee0409/omni-harness-kit/accessibility-checks/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/jhlee0409/omni-harness-kit/accessibility-checks"><img src="https://agentmods.dev/badge/skills/jhlee0409/omni-harness-kit/accessibility-checks.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 | $0.00179 | $0.02914 |
| Opus 5 | $0.00089 | $0.01457 |
| Sonnet 5 | $0.00036 | $0.00583 |
| Haiku 4.5 | $0.00018 | $0.00291 |
Grade A, and why
accessibility-checks 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 10d 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 — 129 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Accessibility checks — WCAG 2.2 AA, measured in a real browser
Accessibility is a MEASURED property of the rendered DOM, never a source-reading guess. aria-label in JSX proves nothing if the runtime node is display:none, the contrast ratio is 3.9:1, or the tap target renders at 32px. This skill runs the audit against the LIVE page with omp's native browser tool and cites the number for every verdict. "looks accessible" is BANNED — every row carries a measured value or it is CANT-VERIFY.
Setup
git diff --name-only→ identify the changed route/component. Read itsDESIGN.md(tokens/tone) if present.- Start the dev server on its own port; wait for a stable anchor before probing (a cold compile is not "down").
browseractionopenthe changed route. Then drive every check with actionrun+tab.evaluate.- All snippets below run inside
tab.evaluate(() => { ... })and RETURN a JSON value — that returned value is the evidence you cite.
1. Semantic structure — landmarks + heading order (WCAG 1.3.1, 2.4.6)
tab.evaluate(() => {
const landmarks = [...document.querySelectorAll('header,nav,main,aside,footer,[role="banner"],[role="navigation"],[role="main"],[role="contentinfo"]')].map(e => e.tagName+ (e.getAttribute('role')?`[${e.getAttribute('role')}]`:''));
const headings = [...document.querySelectorAll('h1,h2,h3,h4,h5,h6')].map(h => ({ lvl:+h.tagName[1], text:h.textContent.trim().slice(0,40) }));
return { mainCount: document.querySelectorAll('main,[role=main]').length, landmarks, headings };
})
FAIL if: mainCount !== 1; no h1; heading levels skip (h2→h4). Cite the actual level sequence.
2. Keyboard nav + focus order + visible focus ring (WCAG 2.1.1, 2.4.3, 2.4.7)
- Enumerate the tab order and confirm it follows DOM/visual order (no positive
tabindexreordering surprises):
tab.evaluate(() => [...document.querySelectorAll('a[href],button,input,select,textarea,[tabindex]')]
.filter(e => !e.disabled && e.offsetParent !== null)
.map(e => ({ tag:e.tagName, ti:e.tabIndex, label:(e.textContent||e.value||e.ariaLabel||'').trim().slice(0,30) })))
FAIL if any positive tabIndex > 0 (manual reorder trap), or a control is reachable by mouse but absent from this list.
- Visible focus ring — focus each interactive node and read
:focus-visiblecomputed style; a ring is only real ifoutline-width/box-shadowactually changes on focus:
tab.evaluate(() => {
const el = document.querySelector('button'); // repeat per control class
el.focus();
const s = getComputedStyle(el);
return { outlineWidth:s.outlineWidth, outlineStyle:s.outlineStyle, boxShadow:s.boxShadow };
})
FAIL if focused control shows outline-style:none AND no distinguishing box-shadow (invisible focus). outline:0 with no replacement = FAIL.
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.
- 10d ago First seen · 129 lines · 179 tokens per session scan A 03418005a7e1
accessibility-checks is a skill published in the GitHub repository jhlee0409/omni-harness-kit (2 stars, last pushed 1mo ago), licensed MIT. It adds 179 tokens to every session and 2,914 once invoked, about $0.0009 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
journey-simulation
Use when caller wants to observe how a stranger encounters a flow, artifact, or sandbox — triggers like "simulate a user journey", "test our onboarding / checkout / signup", "will my ICP convert", "how does a cold reader experience this README", "first-time user test", "cognitive walkthrough", or any request to…
screen-reader-testing
Test web applications with screen readers including VoiceOver, NVDA, and JAWS. Use when validating screen reader compatibility, debugging accessibility issues, or ensuring assistive technology support.
live-preview
Mid-build visual verification loop. Takes screenshots of components during construction, not just after. Catches visual regressions and invisible features before they compound. Requires Playwright or similar screenshot tool.
test-implement
Implements React/TypeScript unit, integration, and browser E2E tests with the repository's configured runner, mocks, setup, and browser harness. Use when creating or completing frontend tests and generated test skeletons.
symfony:e2e-panther-playwright
Write end-to-end tests with Symfony Panther 2.4 for browser automation or Playwright for complex scenarios.
symfony:functional-tests
Write functional tests for Symfony controllers and HTTP endpoints using WebTestCase, getContainer, loginUser, and DAMA rollback.