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 robcsaszar/tend --skill tend-a11ygit clone --depth 1 https://github.com/robcsaszar/tendWrote 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/robcsaszar/tend/tend-a11y)<a href="https://agentmods.dev/skills/robcsaszar/tend/tend-a11y"><img src="https://agentmods.dev/badge/skills/robcsaszar/tend/tend-a11y.svg" alt="Measured on agentmods" 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 | $0.00138 | $0.01739 |
| Opus 5 | $0.00069 | $0.00870 |
| Sonnet 5 | $0.00028 | $0.00348 |
| Haiku 4.5 | $0.00014 | $0.00174 |
Grade A, and why
tend-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 7d 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 — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Tend A11y — the herald
Find the single highest-priority accessibility, UX, or copy-tone issue in this repo's markup, fix it minimally, verify, hand back the diff for review. One issue per run.
0. Load config
- Read
.claude/tend/config.yaml. Useskills.a11y(off-limits globs,notes). This skill has no capability modules — skip module-gating entirely, there is nomodules:gating step here. - No config? Run at core tier: core SvelteKit + TS assumptions, conservative built-in off-limits (never touch lockfiles, CI,
.env, config). Emit once: "runtend-onboardto sharpen." Continue.
1. Triage
MANDATORY READ references/scan-core.md.
- Establish the off-limits set: config's
skills.a11y.off-limitsplus the built-in rails (never touch lockfiles, CI config,.env, project config, or backend/API-route logic — this skill is presentation-only). - Establish entry points: this repo's markup files —
.sveltecomponents,+page.svelte/+layout.svelteroutes, or whatever template files the stack actually uses. Checkskills.a11y.notesfor a named component-library directory (e.g.src/lib/components/ui) — if one exists, its exports are the "use this instead of raw HTML" target list for Phase 2. - Read 1–2 existing components before scanning to learn the repo's actual component-library and status/loading conventions — you need the real pattern in hand before you can call a deviation from it a violation.
Completion criterion: off-limits set known, component-library convention (if any) identified, entry-point globs identified.
2. Scan — priority order, stop at the first real hit
Ordered highest-value (most silent-to-users) first — stop at the first real, evidence-backed hit:
- Raw interactive HTML where a component-library equivalent exists — a raw
<button>/<a>/<input>alongside aButton/Link/Inputexport elsewhere in the repo. - Icon-only or icon-primary interactive elements missing an accessible name — no
aria-label, no visually-hidden text, notitlefallback. <img>withoutalt— a missing attribute is the violation; an intentionally emptyalt=""on a decorative image is not.- Form inputs without an associated label — no
label for=,aria-label, or component-library label prop. Audit the rendered count, not just the template: inputs generated inside a loop ({#each}or equivalent) produce N anonymous controls at runtime. role="button"(or a bareonclickon a non-interactive element) missing the paired keyboard handler —onkeydownfor Enter + Space. Native<button>gets this for free; this only fires on the raw-element escape hatch.- Async success/error/loading feedback rendered as a plain, unannounced element — no
role="alert"/role="status", noaria-live, and no existing component-library banner/toast primitive already used elsewhere in the repo. - A destructive action (delete/remove/reset) with no confirmation step before it takes effect.
- Copy-tone violations, checked against the config-driven allow/deny list (
skills.a11y.notes, or a project tone doc it points to). If no repo-specific list exists, fall back to generic defaults: an imperative softener like "Please" prefacing a validation/error instruction ("Please enter a name" vs. "Enter a name"), or a needless exclamation mark in routine confirmation/instructional copy. Scan both rendered markup text and string literals inside<script>blocks — copy assigned to a variable, passed to a toast/alert call, or set via.textContent/.innerTextlives in script blocks, not just markup, and a markup-only grep will miss the real validation strings.
What ships with it
7 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.
- 7d ago First seen · 83 lines · 138 tokens per session scan A e5fac41a8b3c
tend-a11y is a skill published in the GitHub repository robcsaszar/tend (0 stars, last pushed 4d ago), licensed MIT. It adds 138 tokens to every session and 1,739 once invoked, about $0.0007 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
frontend-review
Frontend Code Review (Angular & React): Expert review of frontend code focusing on Angular, React, component architecture, state management, performance, accessibility, and UI patterns. Use whenever the user wants a review of frontend code, component design, state management, or mentions Angular, React, TypeScript…
scroll-craft
Build premium scroll-driven landing pages for service, product, food, and drink brands. Plan the visitor journey, page grammar, emotional peak, and bespoke signature move. Create dimensional heroes with independent visual planes, restrained motion, and separate mobile composition. Use supplied photos and footage or…
accessibility
Audit and improve web accessibility following WCAG 2.2 guidelines. Use when asked to "improve accessibility", "a11y audit", "WCAG compliance", "screen reader support", "keyboard navigation", or "make accessible".
recipe-front-design
Execute from repository evidence through applicable UI Spec and optional ADR decisions to complete frontend Design Doc approval.
ring:validating-ux-completeness
Validating that UX specifications are complete before technical design: a read-only checklist over wireframes, states, responsive behavior, accessibility, and component-library alignment, emitting a DESIGN VALIDATED / NEEDS REVISION verdict to design-validation.md. Standalone utility — run after a product-designer…
ring:checking-frontend-quality
Checking frontend quality against changed UI via ring:qa-frontend in accessibility, visual, e2e, or performance mode and aggregating pass/fail verdicts. Use when a frontend change needs standalone a11y, visual-snapshot, Playwright e2e, or Lighthouse/Core-Web-Vitals validation outside the dev cycle. Skip for…