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 takaoumehara/superforge-skill --skill superforge-a11ygit clone --depth 1 https://github.com/takaoumehara/superforge-skillWrote 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/takaoumehara/superforge-skill/superforge-a11y)<a href="https://agentmods.dev/skills/takaoumehara/superforge-skill/superforge-a11y"><img src="https://agentmods.dev/badge/skills/takaoumehara/superforge-skill/superforge-a11y/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/takaoumehara/superforge-skill/superforge-a11y"><img src="https://agentmods.dev/badge/skills/takaoumehara/superforge-skill/superforge-a11y.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 | $0.00230 | $0.02324 |
| Opus 5 | $0.00115 | $0.01162 |
| Sonnet 5 | $0.00046 | $0.00465 |
| Haiku 4.5 | $0.00023 | $0.00232 |
Grade A, and why
superforge-a11y 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 — 210 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Superforge A11y — Accessibility Audit & Remediation Engine
Accessibility is not a scanner score. A page can pass every automated rule and still be unusable with a screen reader, unreachable by keyboard, and illegal in the EU. This skill runs the parts a machine cannot run, and says which specific person is blocked by each finding.
0. The rule this skill exists to enforce
Never report a conformance result that came only from an automated tool.
axe-core — the engine inside almost every scanner — ships 63 rules for WCAG 2.0/2.1/2.2 Level A and AA. Level AA has 55 success criteria. The rules are not one-per-criterion: whole criteria (focus order, meaningful sequence, link purpose in context, error suggestion, consistent help, dragging movements, accessible authentication) have no automated rule at all, because passing them is a judgment about meaning.
So: a clean axe run is the start of the audit, and the report must say so.
Anything else is a false clearance the user will act on.
1. Scope before auditing
Fix four things and write them at the top of the report. Guess the defaults and confirm rather than interrogating.
| Default if unstated | |
|---|---|
| Target level | WCAG 2.2 AA — 31 Level A + 24 Level AA criteria |
| Surface | every distinct template/screen, plus each step of the primary flow end to end |
| Platform | infer from the repo (Web / iOS / Android) |
| Legal standard | ask only if the user mentions a market; otherwise audit to 2.2 AA, which meets or exceeds every current regime → references/conformance-and-law.md |
Conformance is claimed for whole pages and complete processes, never for a component. If a checkout is 4 steps and step 3 fails, the process fails. Auditing one component in isolation is fine — calling it conformant is not.
2. The seven passes
Run them in this order. Each one finds what the previous one structurally cannot. Record the evidence per pass; a pass with no evidence did not happen.
What ships with it
11 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.
- README.es.md 6.0 KB
- README.ja.md 6.9 KB
- README.ko.md 6.5 KB
- README.md 5.4 KB
- README.zh-CN.md 5.5 KB
- references/audit-protocol.md 11 KB
- references/conformance-and-law.md 7.1 KB
- references/native-platforms.md 6.7 KB
- references/tooling.md 7.1 KB
- references/wcag22-ledger.md 17 KB
- scripts/contrast.py 9.9 KB runs code
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 · 210 lines · 230 tokens per session scan A e9c946515703
superforge-a11y is a skill published in the GitHub repository takaoumehara/superforge-skill (1 stars, last pushed 25d ago), licensed MIT. It adds 230 tokens to every session and 2,324 once invoked, about $0.0011 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
openai-frontend-design
Use for new frontend applications, dashboards, games, creative websites, hero sections, and visually driven UI from scratch, or when the user explicitly asks for a redesign/restyle/modernization. Builds from clean, airy, high-taste, readable image-generated concept design with section-specific references, faithful…
ultralytics-branding
This skill should be used when creating Ultralytics-branded content of any kind, including PDF, PPTX, Canva, or Google Slides presentations, DOCX documents, marketing or newsletter HTML emails, social media post copy or visuals, website or landing page designs, and dataset or model result visuals. Provides official…
lov-maintain-partners
Maintain the Skill Publisher website's partners section AND align partner logo rows on event posters / hero strips: reuse lov-find-logo for brand logo discovery, normalize collected logos to a 240px-tall content canvas (retina-ready), rasterize SVGs via rsvg-convert before normalizing (so SVG viewBox padding gets…
lov-gen-logo
A logo and application-icon design workflow that creates distinct, versioned drafts and keeps brand marks separate from app icons. It also prepares selected assets for their intended uses, such as websites or device screens.
lov-professional-portrait
Turn a single-person photo into a polished professional portrait while preserving the person's identity. Use restrained skin retouching, exposure cleanup, optional hat removal and hairstyle reconstruction, background polish, and an explicit quality gate. Trigger when the user asks for a professional headshot, business…
lov-find-logo
A tool for collecting company or product logos from public web sources such as a website’s icon or social-preview image.