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 Noibu/ai-plugin --skill accessibility-auditgit clone --depth 1 https://github.com/Noibu/ai-pluginWrote 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/noibu/ai-plugin/accessibility-audit)<a href="https://agentmods.dev/skills/noibu/ai-plugin/accessibility-audit"><img src="https://agentmods.dev/badge/skills/noibu/ai-plugin/accessibility-audit/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/noibu/ai-plugin/accessibility-audit"><img src="https://agentmods.dev/badge/skills/noibu/ai-plugin/accessibility-audit.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 | $0.00079 | $0.04099 |
| Opus 5.5 | $0.00032 | $0.01640 |
| Sonnet 5.5 | $0.00016 | $0.00820 |
| Haiku 4.5 | $0.00008 | $0.00410 |
Grade A, and why
accessibility-audit 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 1mo 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 — 151 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Accessibility Audit
Requires Cowork, linked to a computer with the Claude desktop app open — see Setup below before anything else. This does not work in plain Claude chat, desktop app included.
- On-demand — someone asks for an audit right now. Produces a PDF report, then offers to open PRs/tickets for the top fixable findings immediately, and — once that's resolved — offers to put the whole thing on a recurring schedule, optionally with a digest sent to email/Slack/Notion.
- Scheduled — the same audit, run automatically on a cadence, with any configured fix automation applied before the report is generated. The full PDF stays in that run's session — it is never auto-shared to email, Slack, or Notion, since attaching a generated PDF to those isn't reliable. If a digest was configured, a short exec-summary-plus-links message (not the PDF) goes to the configured channel(s) instead.
Call metadata: on every noibu_* call this skill drives, pass skill: "accessibility-audit" and a short skillStep (see the querying-noibu-data skill).
Setup
The scan needs mcp__remote-devices__Claude_Browser__*, which exists only in a Cowork session linked to a computer running the Claude desktop app — being in the desktop app isn't enough by itself. If those tools are genuinely absent (not just offline): tell the operator plainly this needs Cowork linked to a computer with the desktop app open, and stop. Never offer a reduced fallback (e.g. manual web-fetch review) — it's a materially weaker audit than this skill promises.
Nothing else is required to run a plain audit — it never touches a repo or ticket platform on its own.
- Scanning browser — the built-in browser (
mcp__remote-devices__Claude_Browser__*), for both on-demand and scheduled runs. Once confirmed present (see above), it still needs the linked desktop app to be open and online, not just installed, at the moment it's used. Not available at firing time (scheduled) → report the failure plainly rather than scanning some other way; this means the desktop app wasn't open then, not that anything is broken. Don't substitute Claude in Chrome or the linked-device bridge here — they share the same underlying constraint, not a way around it. - Domain — UUID → use it. Name only →
noibu_get_domain. Neither →noibu_list_domains; use it if there's exactly one, else ask. Never block on this.
What ships with it
5 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.
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.
- 1mo ago Changed · +2 lines 1b5379ceabef
- 1mo ago First seen · 149 lines · 79 tokens per session scan A 022e5f5dcd19
accessibility-audit is a skill published in the GitHub repository Noibu/ai-plugin (5 stars, last pushed 13d ago), licensed MIT. It adds 79 tokens to every session and 4,099 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-09-16.
accessibility-audit compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| accessibility-auditNoibu/ai-plugin | Skill | 5 | 13d ago | MIT | A | 79 | Claude Code |
| shopifyMindrally/skills | Skill | 301 | 1mo ago | Apache-2.0 | A | 27 | Claude Code |
| shopify-theme-optimizationnexscope-ai/eCommerce-Skills | Skill | 1.4k | 2mo ago | MIT | A | 25 | Claude Code, Codex |
| shopify-theme-development-guidelinesMindrally/skills | Skill | 301 | 1mo ago | Apache-2.0 | A | 25 | Claude Code |
| shopify-theme-developmentfinsilabs/awesome-ecommerce-skills | Skill | 63 | 19d ago | MIT | A | 28 | Claude Code, Codex |
| responsive-storefrontfinsilabs/awesome-ecommerce-skills | Skill | 63 | 19d ago | MIT | A | 29 | Claude Code, Codex |
| storefront-themingfinsilabs/awesome-ecommerce-skills | Skill | 63 | 19d ago | MIT | A | 31 | Claude Code, Codex |