claude

synthetic-usability-review

Checks a working service for usability problems before testing with real people. It tries real jobs as the people who use the service, checks every screen against Nielsen's usability heuristics, Microsoft's Guidelines for Human-AI Interaction and the GOV.UK Design System, and writes a plain English report. Not a replacement for research with real people.

2 on GitHubv0.1.0updated 1 month ago
/

Plugin synthetic-usability-review from marketplace defra-ai-plugins in DEFRA/defra-ai-plugins. Commands read from the repo's manifest — 1 month ago.

What it installs

Skills (1)

  • synthetic-usability-review

    Check a working service for usability problems before testing it with real people. Try real jobs in a browser as the people who use it, check each screen against Nielsen's 10 usability heuristics, Microsoft's Guidelines for Human-AI Interaction and the GOV.UK Design System, and write a plain English

View source on GitHub

More from DEFRA/defra-ai-plugins

All 6 plugins →

Builds Defra-compliant frontends following the GOV.UK Design System, WCAG 2.2 AA, and Defra software development standards (Hapi + Nunjucks + SCSS + Vitest).

5 skills · 1 agent · hooks

2

Cross-cutting Defra standards as installable skills and guardrail hooks: branching, commit messages, code-quality gates, security & PII handling, and accessibility. Referenced by every other Defra plugin so cross-cutting standards live in one place. No agent — skills + hooks only.

5 skills · hooks

2

Creates well-structured JIRA tickets (stories and tasks) for any team. Ships default templates for user-facing stories and technical tasks, with support for custom templates at runtime.

2 skills · 1 agent

2

User-centred design skills for Defra: service blueprints and journey maps, document and presentation standards (GOV.UK content design, Defra visual standards), and interaction and content design for CDP frontends. No agent, skills only, built from the AI Capability and Enablement team's delivery work.

3 skills

2

Intercepts all LLM interactions — prompts, tool inputs, and tool outputs — and redacts UK PII before data reaches or leaves the model, using a configurable Python redaction script.

1 skill · hooks

2

Maintain synthetic-usability-review?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex plugins6 plugins