hjemmesidekongen/visual-verification
plugins/smedjen/skills/visual-verification/SKILL.md
Captures Playwright screenshots at 3 breakpoints and verifies UI output against design intent. Three verification tiers: automated layout checks, LLM vision comparison, and human escalation for low-confidence results.
testing
Updated Apr 5, 2026
$ install --global
skillsauthnpx skillsauth add hjemmesidekongen/ai visual-verificationInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
Security Scan Results
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
3
Scanners Passed
9
Scanners in report
Clean
TrivyContainer and dependency vulnerability scanner
95%
Clean
SemgrepStatic code analysis for vulnerabilities
95%
Clean
mcp-scan (Snyk)Model Context Protocol security validation
95%
Skipped
Snyk (dep)Open source security scanning
50%
Skipped
Socket.devSupply chain security analysis
50%
Skipped
VirusTotalMulti-engine malware detection
50%
Skipped
CrowdStrikeAdvanced threat intelligence
50%
Skipped
OSV-ScannerOpen Source Vulnerability database check
50%
Skipped
OWASP Dep-Check
50%
Last scanned: Apr 5, 2026, 5:30 PM13.2s2 files scanned
SKILL.md
- name:
- visual-verification
- description:
- >
- against design intent. Three verification tiers:
- automated layout checks,
- user_invocable:
- false
- interactive:
- false
- model_tier:
- senior
- depends_on:
- []
- type:
- data_validation
- - name:
- no_unresolved_escalations
- verify:
- All human-escalated items have been reviewed or acknowledged
- fail_action:
- Flag unresolved escalations in gate results
- on_fail:
- Visual verification incomplete — capture missing screenshots or resolve escalations
- on_pass:
- Visual output matches intent at all breakpoints
- origin:
- smedjen
- inspired_by:
- agency visual-verification + D-019
- ported_date:
- 2026-03-09
- iteration:
- 1
- changes:
- Built for smedjen with 3-tier verification and Playwright MCP integration
Visual Verification
Screenshots at 3 breakpoints, compared against intent. Only triggered for UI changes.
Breakpoints
| Name | Width | Represents | |------|-------|-----------| | mobile | 375px | iPhone SE | | tablet | 768px | iPad | | desktop | 1280px | Standard laptop |
Trigger Detection
Activated when changed files match: components/, pages/, layouts/,
styles/, *.css, *.scss, *.tsx with JSX, *.vue with template.
Three Verification Tiers
- Automated — Screenshot captured, basic layout assertions (no overflow, no overlapping elements, correct breakpoint width).
- LLM Vision — Compare screenshot to design reference or stated intent. Check color, spacing, typography, alignment against spec.
- Human Escalation — If LLM confidence < 0.7, flag for human review. Provide side-by-side comparison with specific concern noted.
Screenshot Storage
Screenshots saved to .ai/tasks/visual/<task-id>/:
mobile.png,tablet.png,desktop.pngcomparison.yml— verification results per breakpoint
See references/process.md for Playwright config, comparison algorithm, and escalation rules.
Related Skills
hjemmesidekongen/brand-strategy
development
VerifiedTrustedCommunity
Creates a brand from scratch through market research and interactive sparring. Runs competitive research via Perplexity, then guides the user through positioning, audience, voice, values, and content pillars. Produces the full brand guideline set at .ai/brand/{name}/. Use when building a new brand, defining brand strategy for a product, or when /våbenskjold:create is invoked.
SKILL.mdUpdated Apr 5, 2026
hjemmesidekongen/brand-loader
testing
VerifiedTrustedCommunity
Loads brand guidelines from .ai/brand/{name}/ and makes them available to the current context. Progressive disclosure: L1 confirms brand exists, L2 loads summary, L3 loads specific files on demand. Use when a downstream skill or user needs brand context, or when /våbenskjold:apply is invoked.
SKILL.mdUpdated Apr 5, 2026
hjemmesidekongen/brand-evolve
documentation
VerifiedTrustedCommunity
Guided reinvention of an existing brand guideline. Loads current brand from .ai/brand/{name}/, identifies what to keep vs change, and walks the user through targeted evolution. Preserves brand equity while updating positioning, voice, or values. Use when refreshing a brand or when /våbenskjold:evolve is invoked.
SKILL.mdUpdated Apr 5, 2026
hjemmesidekongen/brand-audit
development
VerifiedTrustedCommunity
Codifies an existing brand from materials, samples, and references. Analyzes provided content to extract voice patterns, values, and positioning. Produces the same guideline format as brand-strategy. Use when a brand already exists but isn't documented, or when /våbenskjold:audit is invoked.
SKILL.mdUpdated Apr 5, 2026