Update to Marketing Swarm

{
  "description": "Use when an ad may be fatiguing, the user needs refresh timing, or they want new variants derived from proven creative elements while preserving a clear testing logic.",
  "included_files": [
    {
      "relative_path": "agents/openai.yaml",
      "size_in_bytes": 283
    }
  ],
  "name": "creative-fatigue-mutation",
  "skill_md_contents": "---\nname: creative-fatigue-mutation\ndescription: Use when an ad may be fatiguing, the user needs refresh timing, or they want new variants derived from proven creative elements while preserving a clear testing logic.\n---\n\n# Creative Fatigue and Mutation\n\nDistinguish fatigue from other reasons a creative slows down, then plan refreshes that teach something.\n\n## Fatigue assessment\n\nUseful evidence includes launch date, spend/exposure over time, reach, frequency, CPM, CTR, CPC, CVR, CPA/ROAS, audience size, placement mix, and creative rotation history.\n\nSignals can include:\n\n- rising frequency with falling response\n- CTR decay after stable delivery\n- worsening CPA/ROAS without a corresponding funnel/site change\n- repeated exposure concentrated in a finite audience\n- performance recovery after refresh\n\nDo not infer fatigue from age alone. A creative can be old and still incremental, or new and already weak.\n\nClassify state as:\n\n- `fresh`\n- `performing`\n- `early-fatigue`\n- `fatigued`\n- `exhausted`\n- `uncertain`\n\nUse `uncertain` when attribution, audience, auction, or site changes could explain the decay.\n\n## Mutation logic\n\nStart from the creative hypothesis, not random cosmetic changes. Preserve one proven element and vary one meaningful dimension when possible.\n\nMutation dimensions can include:\n\n- hook\n- angle / problem framing\n- proof mechanism\n- offer framing\n- demo sequence\n- CTA\n- visual opening\n- copy structure\n- format / duration\n\nUse crossover only when two parent creatives have independently supported strengths and the combination has a clear hypothesis.\n\n## Refresh plan\n\nFor each proposed variant include:\n\n- what stays constant\n- what changes\n- why it changes\n- intended audience response\n- success metric\n- minimum evidence needed before judging\n\n## Output contract\n\nReturn fatigue diagnosis, evidence, alternative explanations, refresh priority, and a test-ready variant matrix. Do not promise a remaining lifespan in days unless the time-series evidence supports that estimate.\n"
}