design-review

Run structured design reviews that produce actionable findings and clear approval decisions. Use when a design artifact needs formal review for usability, accessibility, consistency, and implementation readiness before handoff or approval; do not use for backend data-model or deployment pipeline decisions.

Safety Notice

This listing is imported from skills.sh public index metadata. Review upstream SKILL.md and repository scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "design-review" with this command: npx skills add kentoshimizu/sw-agent-skills/kentoshimizu-sw-agent-skills-design-review

Design Review

Overview

Use this skill to convert subjective design feedback into prioritized, evidence-based decisions that unblock implementation.

Scope Boundaries

  • Use this skill when the task matches the trigger condition described in description.
  • Do not use this skill when the primary task falls outside this skill's domain.

Shared References

  • Severity calibration guidance:
    • references/design-review-severity-guidance.md

Templates And Assets

  • Findings template:
    • assets/design-review-findings-template.md
  • Approval decision log:
    • assets/design-review-decision-log-template.md

Inputs To Gather

  • Review target (screens, flows, interaction states) and intended users.
  • Acceptance criteria and non-negotiable constraints.
  • Accessibility and localization expectations for in-scope surfaces.
  • Delivery timeline and risk tolerance for unresolved issues.

Deliverables

  • Findings list with severity, user impact, engineering impact, and owner.
  • Explicit approval decision: approved, conditional approval, or rejected.
  • Remediation plan with due dates and verification method.
  • Residual risk statement for any deferred non-blocker issues.

Quick Example

  • Blocker: keyboard focus order breaks checkout completion.
  • Major: ambiguous copy causes wrong destructive action selection.
  • Minor: visual rhythm inconsistency without measurable task impact.
  • Decision: reject due to blocker; re-review only affected flow after fix.

Quality Standard

  • Findings are reproducible and linked to concrete evidence.
  • Severity reflects impact, not reviewer preference.
  • Blockers are separated from improvements and tracked independently.
  • Approval decision is auditable and tied to explicit criteria.

Workflow

  1. Confirm review scope, criteria, and decision authority.
  2. Evaluate critical user journeys before secondary surfaces.
  3. Log findings with impact, severity, and remediation guidance in assets/design-review-findings-template.md.
  4. Resolve or defer issues using explicit risk acceptance rules and references/design-review-severity-guidance.md.
  5. Publish final decision and follow-up actions in assets/design-review-decision-log-template.md.

Failure Conditions

  • Stop when review scope or acceptance criteria are ambiguous.
  • Stop when critical flows cannot be validated end-to-end.
  • Escalate when blocker findings remain open near implementation handoff.

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

Automation

architecture-clean-architecture

No summary provided by upstream source.

Repository SourceNeeds Review
Automation

information-architecture

No summary provided by upstream source.

Repository SourceNeeds Review
Automation

db-normalization

No summary provided by upstream source.

Repository SourceNeeds Review
Automation

schema-evolution-governance

No summary provided by upstream source.

Repository SourceNeeds Review