CLAUDE.mdA

CLAUDE.md from zivtech/accessibility-skills is a CLAUDE.md file: the standing instructions Claude Code reads in that repo.

Install it, pinned
mdr add zivtech/accessibility-skills/CLAUDE.md@git:20260912.22e662d

Pins git:20260912.22e662d in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 14 of 14 checks passed
Kind
CLAUDE.md file
Repository
zivtech/accessibility-skills
Path
CLAUDE.md
Versions recorded
7
Latest version
git:20260912.22e662d, committed Sep 12, 2026
First version recorded
Sep 4, 2026
Size
17,441 bytes
Repository stars
13
License
GPL-3.0

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# accessibility-skills

This repository contains the standalone Accessibility Skills (accessibility-skills) bundle — planner, critic, tester, and perspective auditor — plus cross-model benchmark assets. Claude Code is one supported install/runtime surface; the evals and runners now compare Claude, Codex/OpenAI, local Ollama models, and other hosted model families as peer baselines.

## Dead Output

**What dead looks like in this repo:**
- Accessibility reviews that pattern-match ARIA attributes without testing whether the interaction actually works for the user. A finding that says "missing aria-label" without explaining what a screen reader user would experience is dead.
- Plans that cite WCAG success criteria by number without engaging with the actual user experience they protect. "Violates 4.1.2" is a citation, not a finding.
- Findings rated CRITICAL because the checklist says so, not because a real user would be blocked. Severity must reflect impact on people, not rule weight.
- Eval fixtures that test whether the skill finds the planted bug without testing whether it avoids false alarms on clean code. A critic that flags everything is dead — it just looks thorough.
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add zivtech/accessibility-skills/CLAUDE.md@sha256:7965c90ab95f4758

The line at the top of this page pins the label git:20260912.22e662d, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_4dqbnera6bjojgt4.svg)](https://markdownregistry.com/a/art_4dqbnera6bjojgt4)

3 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 7 versions of this file, between Sep 4, 2026 and Sep 12, 2026, about one every day.

versioncommittedcommitsizeaudit
git:20260912.22e662d latest2026-09-12 22e662d 17,441 BA view · diff
git:20260912.bc5ae1b2026-09-12 bc5ae1b 16,609 BA view · diff
git:20260904.5d8e7412026-09-04 5d8e741 32,986 BA view · diff
git:20260904.57034be2026-09-04 57034be 32,009 BA view · diff
git:20260904.43300bd2026-09-04 43300bd 31,799 BA view · diff
git:20260904.4fd73902026-09-04 4fd7390 31,269 BA view · diff
git:20260904.2448d812026-09-04 2448d81 29,743 BA view

Audit of the latest version

A  14 of 14 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Size between 200 bytes and 200 KB (17441 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

zivtech/accessibility-skills · 13 stars · license GPL-3.0 · pushed 2026-10-23 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_4dqbnera6bjojgt4
GET https://markdownregistry.com/api/v1/resolve?ref=zivtech/accessibility-skills/CLAUDE.md
GET https://markdownregistry.com/api/v1/blob/7965c90ab95f47583423d558ebb9ddfdb944813ddbe4058c869c2c35acdee1e5

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: what a CLAUDE.md file is · keeping AGENTS.md and CLAUDE.md in sync · which agent file does what.

More from zivtech/accessibility-skills

a11y-content-judgment SKILL.md skill
zivtech/accessibility-skills · .agents/skills/a11y-content-judgment/SKILL.md · Load this skill when an audit needs the judgment-shaped WCAG criteria that scanners cannot decide — are page titles…
v0.1.0 · audit A · 13 stars
a11y-critic SKILL.md skill
zivtech/accessibility-skills · .agents/skills/a11y-critic/SKILL.md · Use when you have an existing component, flow, or interface and need an evidence-backed accessibility design review…
v1.1.1 · audit A · 13 stars
a11y-planner SKILL.md skill
zivtech/accessibility-skills · .agents/skills/a11y-planner/SKILL.md · Use when you know what component, flow, or interface you need but not yet the right accessibility approach. Best for…
v1.1.0 · audit A · 13 stars
a11y-role-audit SKILL.md skill
zivtech/accessibility-skills · .agents/skills/a11y-role-audit/SKILL.md · ARRM-based accessibility role audit — responsibility-based review lenses
git:20260719.7d69d80 · audit A · 13 stars
a11y-test SKILL.md skill
zivtech/accessibility-skills · .agents/skills/a11y-test/SKILL.md · Use when you need to run real accessibility tests — Playwright keyboard interactions, axe-core scanning, visual…
v1.4.0 · audit A · 13 stars
acr-reporting SKILL.md skill
zivtech/accessibility-skills · .agents/skills/acr-reporting/SKILL.md · Load this skill whenever you are converting a finished audit-scope accessibility evaluation into an Accessibility…
v0.1.0 · audit A · 13 stars
bug-reporting SKILL.md skill
zivtech/accessibility-skills · .agents/skills/bug-reporting/SKILL.md · Load this skill whenever you are filing, reviewing, or generating accessibility bug reports — whether from automated…
v1.0.0 · audit A · 13 stars
maintain-accessibility-skills SKILL.md skill
zivtech/accessibility-skills · .agents/skills/maintain-accessibility-skills/SKILL.md · Safely maintain the public accessibility-skills repository. Use when adding tracked-file hygiene scanners, proving…
git:20260912.bc5ae1b · audit A · 13 stars
perspective-audit SKILL.md skill
zivtech/accessibility-skills · .agents/skills/perspective-audit/SKILL.md · Deep accessibility review from 7 access perspectives — activated by escalation from a11y-planner or a11y-critic when…
v1.0.1 · audit A · 13 stars
a11y-content-judgment SKILL.md skill
zivtech/accessibility-skills · .claude/skills/a11y-content-judgment/SKILL.md · Load this skill when an audit needs the judgment-shaped WCAG criteria that scanners cannot decide — are page titles…
v0.1.0 · audit A · 13 stars
a11y-critic SKILL.md skill
zivtech/accessibility-skills · .claude/skills/a11y-critic/SKILL.md · Use when you have an existing component, flow, or interface and need an evidence-backed accessibility design review…
v1.1.1 · audit A · 13 stars
a11y-planner SKILL.md skill
zivtech/accessibility-skills · .claude/skills/a11y-planner/SKILL.md · Use when you know what component, flow, or interface you need but not yet the right accessibility approach. Best for…
v1.1.0 · audit A · 13 stars

Every file in zivtech/accessibility-skills

Other files named CLAUDE.md

CLAUDE.md CLAUDE.md file
affaan-m/ecc · CLAUDE.md
git:20261020.9471ef4 · audit A · 299,696 stars
CLAUDE.md CLAUDE.md file
firecrawl/firecrawl · CLAUDE.md
git:20261008.5a75dca · audit A · 199,595 stars
CLAUDE.md CLAUDE.md file
garrytan/gstack · CLAUDE.md
git:20261107.ef99eb5 · audit A · 140,694 stars
CLAUDE.md CLAUDE.md file
juliusbrussee/caveman · CLAUDE.md
git:20261106.af560e1 · audit B · 118,366 stars
CLAUDE.md CLAUDE.md file
addyosmani/agent-skills · CLAUDE.md
git:20260720.8fd2081 · audit A · 113,383 stars
CLAUDE.md CLAUDE.md file
nexu-io/open-design · CLAUDE.md
git:20260430.9b57c22 · audit B · 106,541 stars
CLAUDE.md CLAUDE.md file
pbakaus/impeccable · CLAUDE.md
git:20261105.71dfd50 · audit A · 106,186 stars
CLAUDE.md CLAUDE.md file
debpalash/voicestudio · CLAUDE.md
git:20261104.c2998f6 · audit A · 102,675 stars

Browse by kind, by grade A, or by owner.