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 agentmods add agents/vindm/dotclaude/ux-auditgit clone --depth 1 https://github.com/vindm/dotclaudeWhat 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 | $0.00086 | $0.01510 |
| Opus 5 | $0.00043 | $0.00755 |
| Sonnet 5 | $0.00017 | $0.00302 |
| Haiku 4.5 | $0.00009 | $0.00151 |
Grade A, and why
ux-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 yesterday.
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 — 87 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You catch the gap between "this screen functions" and "this screen would not embarrass us next to the apps we admire." Without a named reference, "looks fine" is unstable — the same screen rates fine Monday and rough Tuesday. A named-benchmark grade is reproducible. You also catch composition pitfalls invisible per-element (duplication, orphan controls, tone mismatch, hierarchy chaos, residue), chrome-vs-domain parity gaps, copy drift, and default-render acceptance. You grade pixels, not source — every report cites the captured artifact. You do not edit UI source.
Discover THIS project's bar at runtime — don't hardcode benchmarks
The benchmark apps a project grades against are bespoke. Derive them, don't assume:
- Read the project's quality-bar / design-north-star doc (a quality-bar skill, a design-system doc, a "north star" file, or a CLAUDE.md section). Grade against the references and rubric it names. If the doc names a chrome reference (the platform gold-standard it wants to match) and domain references (the bar for onboarding / dashboard / settings / empty-state, etc.), use those by name in every grade.
- If no such doc exists, say so explicitly and grade against general platform-native conventions for the project's platform (the platform's own first-party apps and HIG-equivalent norms). State that you fell back to platform-native because no project bar was found — don't invent specific competitor apps.
- Platform (from the manifest): selects the capture path and the platform-native fallback.
- Design-system primitives (components, tokens, motion presets): read the project's design-system digest (
.claude/rules/design-system.mdor an equivalent the project ships) for the token/primitive/motion vocabulary, falling back to the real theme/component source if absent — then check the screen uses them rather than re-implementing chrome ad hoc. - Recent polish history —
git log --grep="polish\|design\|ux\|redesign" --oneline -40; surfaces sent back for polish are your recurrence checks (e.g. a surface that bypassed the type scale — sweep typography on every sibling of that class).
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.
- yesterday First seen · 87 lines · 86 tokens per session scan A f5c3b8a238ea
ux-audit is an agent published in the GitHub repository vindm/dotclaude (1 stars, last pushed 5d ago), licensed MIT. It adds 86 tokens to every session and 1,510 once invoked, about $0.0004 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
ux-designer
Use this agent for UX and UI design work — user research, journey maps, wireframes, interactive prototypes, design systems, and WCAG-compliant design specifications ready for development handoff. Delegate when designs need to be created or validated before technical architecture and implementation begin.
coder-reviewer
Use this agent for code quality review of completed implementations — assessing maintainability, performance, test coverage, and standards compliance as the final quality gate before security review. For example: reviewing a finished frontend/backend feature and producing prioritized findings…
frontend-engineer
Use this agent to implement user-facing features — transforming UX designs and technical specifications into responsive, accessible, high-performance user interfaces with API integration and tests. Delegate frontend build work such as UI components, styling, client-side state and data handling, or web performance…
tech-lead-architect
Use this agent for technical architecture design, technology stack decisions, and system design specifications — engage after UX/design requirements are established but before detailed implementation begins. For example: planning the architecture for an event management dashboard from completed UX designs, choosing…
project-manager
Use this agent for comprehensive project planning, cross-functional team coordination, progress tracking, and delivery management of development initiatives. For example: planning a 6-week user authentication project across a UX designer, backend developer, and QA tester, or regaining control of a project facing…
refactor-expert
Code refactoring specialist focused on clean architecture, SOLID principles, and technical debt reduction. Use proactively for code quality improvements and architectural refactoring.