OpenRisk UX Doctrine

SkillMedia

OpenRisk UX doctrine — the four personas, the 15 non-negotiable rules and the 20-criterion grading grid used to design and audit every screen. Load before designing, implementing or auditing any interface.

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the OpenRisk UX Doctrine skill

What this skill tells your AI

The instructions your AI receives, as published by opendefender/openrisk in .claude/skills/openrisk-ux-doctrine/SKILL.md and read by ahel’s review.

The four personas

PersonaJob to be doneFailure mode to avoid
Risk ManagerMaintain the register, run assessmentsData-entry friction, lost work
Compliance OfficerMap controls to frameworks, prove coverageUntraceable evidence
CISOSee exposure at a glance, decide where to spendDashboards hiding the trend
AuditorVerify, sample, export, citeAnything not exportable or timestamped

The 15 rules

  1. Every screen answers: where am I, what can I do, what happened last.
  2. Routine deletion is soft with a 5s undo. Vital deletion is informed friction with an impact readout and a safer alternative offered as a first-class button.
  3. Nothing important behind hover alone on a touch-capable viewport.
  4. Any list over 50 rows: bulk actions, server pagination, export.
  5. Filters persist and are URL-shareable.
  6. Loading, empty, partial, error and permission-denied are all designed.
  7. Error copy: what happened, why, what to do next. Never "an error occurred".
  8. Keyboard-operable end to end. Escape closes. Focus returns to the trigger.
  9. Focus always visible, never trapped.
  10. Severity never encoded by color alone.
  11. Contrast 4.5:1 text, 3:1 UI. Measured.
  12. No layout shift after load. CLS budget 0.05.
  13. Motion honours prefers-reduced-motion; transform and opacity only.
  14. FR and EN both native. No untranslated string reaches a user.
  15. The interface never claims a capability the backend does not provide.

The 20-criterion grid — score 1–5, anything under 4 gets a written fix

Clarity of purpose · Information hierarchy · Navigation predictability · Data density · Scan-ability · Action discoverability · Feedback latency · Error prevention · Error recovery · State completeness · Keyboard operability · Screen-reader semantics · Color independence · Contrast compliance · Responsive integrity · Motion restraint · i18n completeness · Design-system consistency · Performance perception · Product truthfulness.

Signals

GitHub stars
33
Forks
3
Last commit
Sep 2026
Advanced
Catalog kind
skill
Gateway key
openrisk-ux-doctrine
Source
github.com/opendefender/openrisk