accessibility-reviewer

accessibility-reviewer is an agent for Claude Code from RBraga01/builder-design. It costs 52 tokens per session (1,194 once invoked), scanned A, original, MIT.

A review process for checking whether an AI feature can be used by people with disabilities. It examines screen-reader announcements, keyboard use, reading order, cognitive load, and changing AI output.

In plain words
What is it for?
It is for auditing AI interfaces before release and reporting blocking accessibility problems, warnings, and notes.
Why use it?
AI screens change while they are being used, so accessibility problems can be missed by checks designed only for fixed pages.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: model in frontmatter.

Part of the builder-design plugin — 8 skills, 5 agents shipped together

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.

agentmods
npx agentmods add agents/rbraga01/builder-design/accessibility-reviewer
Clone the repo
git clone --depth 1 https://github.com/RBraga01/builder-design

Made for: Claude Code.

Or install builder-design, the plugin that ships this one along with the rest of its 8 skills, 5 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/rbraga01/builder-design/accessibility-reviewer.svg)](https://agentmods.dev/agents/rbraga01/builder-design/accessibility-reviewer)
Your own site
<a href="https://agentmods.dev/agents/rbraga01/builder-design/accessibility-reviewer"><img src="https://agentmods.dev/badge/agents/rbraga01/builder-design/accessibility-reviewer.svg" alt="Measured on agentmods" height="20"></a>
Per session 52 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,194 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00052 $0.01194
Opus 5 $0.00026 $0.00597
Sonnet 5 $0.00010 $0.00239
Haiku 4.5 $0.00005 $0.00119

Measured 5d ago against content hash 272e9bb08d02, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, 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 5d 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.

.claude/agents/accessibility-reviewer.md · 133 lines

How it starts

The opening of the file, as written. The whole thing — 133 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are an accessibility reviewer for AI product interfaces. AI output has specific accessibility requirements that static content does not — streaming text, dynamic state changes, uncertainty signals, agent action timelines, and long-form rendered markdown all require careful ARIA and DOM design.

What You Do

You audit against 5 accessibility concern areas. Every finding is BLOCK, WARN, or NOTE.

Area 1 — ARIA Live Regions

Streaming output container must have:

<div role="log" aria-live="polite" aria-atomic="false">
  • role="log" — correct semantics for a running stream of content
  • aria-live="polite" — announces after user action completes, not mid-keystroke
  • aria-atomic="false" — announces only new content, not the whole region on each token

Error containers must have:

<div role="alert" aria-live="assertive">
  • role="alert" for errors that require immediate attention
  • aria-live="assertive" — interrupts the screen reader (appropriate for errors only)

Streaming cursor must have:

<span class="streaming-cursor" aria-hidden="true">▋</span>
  • aria-hidden="true" — the cursor is decorative; announcing it on every token is unusable

BLOCK if any of these are missing.

Area 2 — Reading Order

DOM order must match the visual reading order. AI interfaces commonly break this when:

  • Feedback buttons (👍 👎) appear in the DOM before the response text (positioned via CSS)
  • Citation footnotes are in a sidebar that appears visually after the response but DOM-before
  • The streaming cursor is injected at the start of the container instead of the end

Check:

  • Tab order through the response container follows top-to-bottom, left-to-right
  • Citation numbers in the response link to the correct reference in the list below
  • The prompt input is the last interactive element in reading order (user returns to it after reading)

Area 3 — Cognitive Load

Truncated content must announce itself:

<button aria-label="Show all 15 items, currently showing 5">Show all</button>

Read the full file on GitHub · 133 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. 5d ago First seen · 133 lines · 52 tokens per session scan A 272e9bb08d02

Subscribe to this mod's changes

accessibility-reviewer is an agent published in the GitHub repository RBraga01/builder-design (2 stars, last pushed 2mo ago), licensed MIT. It adds 52 tokens to every session and 1,194 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.