Update to Designly

{
  "name": "visual-qa",
  "description": "Independent visual quality assurance, hard-gate auditor, and targeted revision router. This skill should be used when reviewing generated or edited visuals, testing hierarchy/craft/brand fidelity, detecting AI slop, checking bounded-edit collateral drift, issuing approval verdicts, or routing targeted RevisionRequests.",
  "included_files": [
    {
      "relative_path": "agents/openai.yaml",
      "size_in_bytes": 380
    },
    {
      "relative_path": "assets/large-logo.svg",
      "size_in_bytes": 663
    },
    {
      "relative_path": "assets/small-logo.svg",
      "size_in_bytes": 663
    },
    {
      "relative_path": "assets/visual-review.template.json",
      "size_in_bytes": 891
    },
    {
      "relative_path": "schemas/visual-review.schema.json",
      "size_in_bytes": 2500
    },
    {
      "relative_path": "scripts/score_review.py",
      "size_in_bytes": 3530
    },
    {
      "relative_path": "scripts/test_gates.py",
      "size_in_bytes": 2619
    }
  ],
  "skill_md_contents": "---\nname: visual-qa\ndescription: Independent visual quality assurance, hard-gate auditor, and targeted revision router. This skill should be used when reviewing generated or edited visuals, testing hierarchy/craft/brand fidelity, detecting AI slop, checking bounded-edit collateral drift, issuing approval verdicts, or routing targeted RevisionRequests.\n---\n\n# Visual QA\n\nReview the actual visual independently. Do not defend upstream decisions and do not approve an uninspected hypothetical output.\n\n## Core review\n\nEvaluate applicable categories including:\n\n- brief accuracy and primary message\n- concept strength\n- hierarchy and composition\n- grouping, spacing and crop\n- typography and exact copy\n- color/contrast\n- brand fidelity\n- product fidelity\n- physical believability\n- cultural/platform fit and inclusive representation\n- overall craft\n- AI slop\n\nWeighted score must meet the configured threshold and every applicable category floor must pass\n\n## Hard gates\n\nFail regardless of average when applicable:\n\n- required copy is wrong\n- Arabic glyphs/connections or RTL behavior are malformed\n- official logo/mark is malformed\n- supplied product identity/proportions materially drift\n- focal anatomy, human dignity, or physical interaction has a critical defect\n- protected edit content changes materially outside what boundary blending requires\n- local edit changes crop, canvas, camera, layout, lighting, text, identity or style that the EditContract locked\n- annotation was applied to the wrong semantic target\n- AI-slop veto triggers\n\n## Bounded-edit review\n\nWhen `edit_state` exists, compare output to the approved `source_checkpoint`, not to a previous failed edit. Evaluate target accuracy, edit-scope accuracy, and collateral change.\n\n## Perception checks\n\n- one-second hierarchy test\n- thumbnail test\n- squint/blur value-mass test\n- grayscale hierarchy test\n- edge/tangency/crop test\n- brand-off specificity test\n- 7-point inclusive representation review\n- effect-subtraction test\n- physics pass\n- character-by-character copy pass\n- source-vs-output edit-scope pass\n\n## AI slop veto\n\nBlock on:\n- any critical finding\n- 2+ major findings\n- 4+ minor findings\n- cumulative pressure >= 6, minor=1 and major=3\n\n## Revision routing\n\nReturn one `RevisionRequest` to the smallest responsible node:\n- concept/originality -> `creative-director`\n- insight depth -> `insight-mining`\n- pattern saturation -> `campaign-canon`\n- activation mechanic -> `brand-activation`\n- narrative arc -> `visual-storytelling`\n- strategy/message -> `creative-strategy`\n- hierarchy/composition -> `composition-director`\n- typography -> `typography-director`\n- Arabic -> `arabic-rtl-director`\n- brand/product identity -> `brand-intelligence`\n- physical integration -> `manipulation-director`\n- wrong annotation target -> `edit-sanitizer` with `annotation_mapping`\n- over-broad edit scope -> `edit-sanitizer` with `edit_scope`\n- collateral drift -> `edit-sanitizer` with `collateral_change`\n- provider instruction mismatch -> `prompt-compiler`\n\n---\n\n## Cross-Skill Neural Connections & References\n\n### Peer & Specialist Skills\n- [Designly Director](../designly-director/SKILL.md) — Lead orchestrator and final signoff authority\n- [Composition Director](../composition-director/SKILL.md) — Hierarchy & spatial layout revisions\n- [Brand Intelligence](../brand-intelligence/SKILL.md) — Brand guidelines & product fidelity revisions\n- [Edit Sanitizer](../edit-sanitizer/SKILL.md) — Bounded edit scope & annotation corrections\n- [Prompt Compiler](../prompt-compiler/SKILL.md) — Provider prompt adjustments\n\n### Schemas & References\n- [Visual Review Schema](schemas/visual-review.schema.json) — Local review schema\n- [Revision Request](../../shared/contracts/revision-request.schema.json) — Mesh revision contract\n- [Edit Contract](../../shared/contracts/edit-contract.schema.json) — Bounded edit contract\n- [Visual QA & Revisions](../../shared/references/visual-qa-and-revisions.md) — Review methodology\n- [AI Slop Taxonomy](../../shared/references/ai-slop-taxonomy.md) — Slop families & severity criteria\n- [Inclusive Representation](../../shared/references/inclusive-representation-and-ethics.md) — 7-point representation checklist\n- [Software Architecture](../../shared/references/software-architecture-and-contracts.md) — Bounded contexts & lock hierarchy\n"
}