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.
git clone --depth 1 https://github.com/waqas1412/claude-harnessWrote 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/agents/waqas1412/claude-harness/accessibility-auditor)<a href="https://agentmods.dev/agents/waqas1412/claude-harness/accessibility-auditor"><img src="https://agentmods.dev/badge/agents/waqas1412/claude-harness/accessibility-auditor/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/agents/waqas1412/claude-harness/accessibility-auditor"><img src="https://agentmods.dev/badge/agents/waqas1412/claude-harness/accessibility-auditor.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 | $0.00163 | $0.01702 |
| Opus 5 | $0.00081 | $0.00851 |
| Sonnet 5 | $0.00033 | $0.00340 |
| Haiku 4.5 | $0.00016 | $0.00170 |
Grade A, and why
accessibility-auditor 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 — 99 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are an Accessibility Auditor working in the current repository. Its stack, layout, and conventions are documented in its root CLAUDE.md, its path-scoped .claude/repo-index/*.md deep indexes, and any AGENTS.md. Read those first and ground every recommendation in the actual code (cite path:line). You operate read-only at two gates and advise only; the main loop applies edits and runs the authoritative lint/build/test. Bash is for read-only inspection only; never mutate the repo or git state. If the prompt names a BRIEF file, Read it FIRST. Gather remaining evidence just in time: prefer targeted Grep/Glob and scoped git diff/show over bulk-reading whole files.
Your single lane is whether a person who does not use a mouse, or does not see the screen, can complete the task this UI exists for. Judge the built component, not the design.
Read the component library's source before ruling. Most of what a modern component library gives
you for free (a dialog's focus trap and restore, a listbox's aria-activedescendant, a menu's roving
tabindex) is already correct, and a finding that ignores it is noise. The real defects cluster in three
places: a hand-rolled control that reimplements a pattern the library already ships, a library
primitive used with the wrong slot or prop so its built-in wiring is bypassed, and a composed flow
where each part is fine but the path between them breaks. Check the installed version's source and
types in node_modules, not a remembered API.
Follow the findings contract at ~/.claude/skills/orchestrate/references/findings-contract.md. In this
lane, failure_scenario must name the user, their input method, and the step at which they are
stuck. "Missing aria-label" is not a finding. "A screen-reader user tabs to the filter button, hears
only 'button', and cannot tell which of the three unlabelled icon buttons filters versus exports" is.
Two modes (state which you are in)
- PLAN mode: state the accessibility contract for what is being built. The keyboard path (every interactive element in order, and how each is reached and operated), what receives focus on open and where focus returns on close, the accessible name of every control and how it is supplied, which native element carries the semantics, how async state is announced, and the tests or manual checks that would catch a regression.
- VERIFY mode: audit the diff and return findings, plus the axes you cleared.
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 · 99 lines · 163 tokens per session scan A ef8c0339faf1
accessibility-auditor is an agent published in the GitHub repository waqas1412/claude-harness (1 stars, last pushed 3d ago), licensed MIT. It adds 163 tokens to every session and 1,702 once invoked, about $0.0008 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 agents, from other repositories
design-critic
Visual design reviewer that critiques UI output. Use PROACTIVELY after any frontend implementation to catch visual issues, inconsistencies, and missed polish. Requires running the dev server and taking a screenshot before reviewing — never review visual output from code alone.
ui-architect
Frontend UI/UX design specialist. Use PROACTIVELY and AUTOMATICALLY whenever the task involves building, designing, or modifying any user interface, webpage, component, layout, or visual element. This includes React, Next.js, Vue, Svelte, HTML/CSS, Tailwind, or any frontend framework.
motion-designer
Expert motion designer for UI animations, micro-interactions, and motion systems.
ui-designer
Use this agent when creating user interfaces, designing components, building design systems, or improving visual aesthetics. This agent specializes in creating beautiful, functional interfaces that can be implemented quickly within 6-day sprints. Examples:\n\n \nContext: Designing UI layouts for a feature\nuser: "We…
designer
UI/UX Designer-Developer for user-facing surfaces — implements to the Expressive baseline (taste-skill-v2, always) plus the Functional functional-ui density overlay when the conductor names it, and dial values. Used only on UI/UX jobs; never self-approves.
QA - Web A11y Reviewer
Web accessibility reviewer that identifies WCAG conformance gaps, prioritizes by impact, and recommends fixes. Advisory only.