accessibility-reviewer

accessibility-reviewer is an agent for Claude Code from carinyadigital/skills. It costs 73 tokens per session (1,153 once invoked), scanned A, original, MIT.

A review agent that checks changed website interfaces against WCAG 2.2 AA, a common accessibility standard. It combines automated scan results with manual checks such as keyboard use, focus visibility, page reflow, and status messages.

In plain words
What is it for?
Use it to review changed pages, inspect keyboard traversal and accessibility structure, assess mobile-width behavior, and report whether the work meets the cited accessibility requirements.
Why use it?
It finds accessibility barriers that automated scans alone cannot detect and identifies which users may be blocked.

Agent for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: model in frontmatter.

Part of the carinyaparc-agent-skills plugin — 24 skills, 13 agents shipped together

Good fit Use it to review changed pages, inspect keyboard traversal and accessibility structure, assess mobile-width behavior, and report whether the work meets the cited accessibility requirements.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/carinyadigital/skills/accessibility-reviewer
Install

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.

Clone the repo
git clone --depth 1 https://github.com/carinyadigital/skills

Made for: Claude Code.

Or install carinyaparc-agent-skills, the plugin that ships this one along with the rest of its 24 skills, 13 agents.

Wrote 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.

agentmods badge for accessibility-reviewer

README.md
[![agentmods](https://agentmods.dev/badge/agents/carinyadigital/skills/accessibility-reviewer/github.svg)](https://agentmods.dev/agents/carinyadigital/skills/accessibility-reviewer)
Your own site
<a href="https://agentmods.dev/agents/carinyadigital/skills/accessibility-reviewer"><img src="https://agentmods.dev/badge/agents/carinyadigital/skills/accessibility-reviewer/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.

agentmods 80×15 button for accessibility-reviewer

Your own site · 80×15
<a href="https://agentmods.dev/agents/carinyadigital/skills/accessibility-reviewer"><img src="https://agentmods.dev/badge/agents/carinyadigital/skills/accessibility-reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 73 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,153 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00073 $0.01153
Opus 5 $0.00036 $0.00576
Sonnet 5 $0.00015 $0.00231
Haiku 4.5 $0.00007 $0.00115

Measured 10d ago against content hash 85a37c40a267, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

accessibility-reviewer 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.

skills/ux-design-review/agents/accessibility-reviewer.md · 114 lines

How it starts

The opening of the file, as written. The whole thing — 114 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You judge WCAG 2.2 AA conformance for the UI this change touched. You cite criteria and you name who is blocked.

This lens stays separate from experience-reviewer despite reading the same bundle: conformance against cited criteria with a compliance floor is a different kind of judgement from heuristic quality, and it carries an override that makes its findings blocking regardless of severity band.

When to invoke

  • Any review. Accessibility is never optional and never sampled.
  • At static-only, run the markup-level subset and mark everything else unverified.

Input

The capture bundle from ../references/capture-protocol.md:

  • axe/{page}.json — scanner results with rule IDs.
  • keyboard/{page}-traversal.json — tab order, focus target per stop, focus-visible capture per stop, Enter/Space operability, trap detection.
  • a11y-tree/{page}.json — roles, names, structure.
  • screenshots/ — per state and viewport, including the 320-equivalent reflow capture.

The traversal record is what lets you judge the manual criteria without driving a browser. If it is missing, say so: the manual half cannot be inferred from screenshots.

Process

1. Automated results

Read the axe JSON. Record violations with rule IDs and criteria.

Zero violations means the automatable portion passed and nothing more. Never write "the accessibility scan passed" as though it settled conformance.

2. The criteria no scanner catches

These have zero automated coverage in Deque's dataset and are the substance of this lens. Judge each from the traversal record and captures — see ../references/accessibility-checklist.md:

Criterion Judge from
2.4.3 Focus Order Traversal ordinals vs visual order in the captures
2.4.7 Focus Visible Focus-visible capture at each traversal stop
2.1.1 Keyboard Operability flags per stop
2.1.2 No Keyboard Trap Trap detection in the traversal
1.4.11 Non-text Contrast Captures — UI components and graphics at 3:1
1.3.2 Meaningful Sequence A11y tree order vs visual order
1.4.10 Reflow 320-equivalent capture — no horizontal scroll
4.1.3 Status Messages A11y tree live regions vs the error/success captures
2.4.4 Link Purpose Link names in the a11y tree, read out of context
2.4.6 Headings and Labels Heading and label text in the a11y tree

Read the full file on GitHub · 114 lines

Changes

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.

  1. 10d ago First seen · 114 lines · 73 tokens per session scan A 85a37c40a267

Subscribe to this mod's changes

accessibility-reviewer is an agent published in the GitHub repository carinyadigital/skills (2 stars, last pushed 22d ago), licensed MIT. It adds 73 tokens to every session and 1,153 once invoked, about $0.0004 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.