ux-audit

An interaction-focused review of a live web app performed as a real user. It tests typing, clicking, sending, and observing across defined scenarios, while checking console errors, failed network requests, layout collapse, and visual quality.

In plain words
What is it for?
Use it to audit a whole app or a scoped flow from a chosen user’s perspective. It produces a verdict, identifies the top five issues, critiques the findings, and applies the smallest suitable fix.
Why use it?
It finds problems that a static code or page scan can miss, including broken interactions, errors, overloaded layouts, and unfinished details.

Command

Install

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.

agentmods
npx agentmods add commands/jezweb/claude-skills/ux-audit
Clone the repo
git clone --depth 1 https://github.com/jezweb/claude-skills
Per session 101 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 358 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00101 $0.00358
Opus 5 $0.00051 $0.00179
Sonnet 5 $0.00020 $0.00072
Haiku 4.5 $0.00010 $0.00036

Measured yesterday against content hash 8d8784c8e6c9, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

ux-audit scanned grade A with 1 finding 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.

Unrestricted tool accesslowExcessive agency

A wildcard tool grant or "run any command" leaves no least-privilege boundary at all.

allowed-tools: "*"

Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.

plugins/dev-tools/commands/ux-audit.md · 22 lines

What it actually says

Load the ux-audit skill and run the audit.

The audit is interaction-first. It cannot produce a verdict without an Interaction Manifest proving real typing, clicking, sending, observing. A static DOM sweep terminates with verdict Incomplete.

Hard gates (cannot be downgraded): console errors / warnings = 0, network 5xx = 0, layout collapse at any tested viewport = 0. A console warning is High minimum. A 5xx is Critical automatically.

If $ARGUMENTS describes a persona ("as a busy broker", "first-time user"), lock it. If $ARGUMENTS scopes the audit ("the dashboard", "just the settings flow"), scope to that area — still exhaustive within the scope.

URL is auto-detected from wrangler.jsonc or running dev server.

Examples:

  • /ux-audit
  • /ux-audit as a time-poor client logging in for the first time
  • /ux-audit the billing flow
  • /ux-audit as an SME owner — multi-pane stress focus
Changes

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.

  1. yesterday First seen · 22 lines · 101 tokens per session scan A 8d8784c8e6c9

Subscribe to this mod's changes

ux-audit is a command published in the GitHub repository jezweb/claude-skills (982 stars, last pushed 2mo ago), licensed MIT. It adds 101 tokens to every session and 358 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 1 finding (unrestricted tool access). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.