skills/visual-verdict/SKILL.md
Structured visual QA verdict for screenshot-to-reference comparisons
npx skillsauth add RobinNorberg/oh-my-copilot visual-verdictInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
<Use_When>
<Output_Contract> Return JSON only with this exact shape:
{
"score": 0,
"verdict": "revise",
"category_match": false,
"differences": ["..."],
"suggestions": ["..."],
"reasoning": "short explanation"
}
Rules:
score: integer 0-100verdict: short status (pass, revise, or fail)category_match: true when the generated screenshot matches the intended UI category/styledifferences[]: concrete visual mismatches (layout, spacing, typography, colors, hierarchy)suggestions[]: actionable next edits tied to the differencesreasoning: 1-2 sentence summary<Threshold_And_Loop>
score < 90, continue editing and rerun /oh-my-copilot:visual-verdict before any further visual review pass.<Debug_Visualization> When mismatch diagnosis is hard:
$visual-verdict as the authoritative decision.differences[] and suggestions[] updates.
</Debug_Visualization>Task: {{ARGUMENTS}}
tools
Durable multi-goal workflow that persists plan/ledger artifacts under .omcp/ultragoal and prints Claude /goal handoff text for the active session
tools
testing
Evidence-driven tracing lane that orchestrates competing tracer hypotheses in Claude built-in team mode
tools
Consensus planning entrypoint that auto-gates vague ralph/autopilot/team requests before execution