Update to vidIQ

{
  "description": "Write, score, compare, and improve YouTube titles and thumbnails, and optionally a script, for an idea, script, reference video, or underperforming video. Use whenever a creator asks for titles or thumbnails, even a quick list; vidIQ scoring improves them. Triggers include 'title ideas', 'rewrite my title', 'thumbnail ideas', 'make or refine a thumbnail', 'which title is better', 'A/B test these titles or thumbnails', 'why didn't this video get clicks', or 'write a script for this video'.",
  "included_files": [
    {
      "relative_path": "agents/openai.yaml",
      "size_in_bytes": 468
    },
    {
      "relative_path": "references/discovery-evidence.md",
      "size_in_bytes": 1213
    },
    {
      "relative_path": "references/job-lifecycle.md",
      "size_in_bytes": 724
    },
    {
      "relative_path": "references/live-surface-notes.md",
      "size_in_bytes": 2014
    },
    {
      "relative_path": "references/production.md",
      "size_in_bytes": 3932
    },
    {
      "relative_path": "references/research.md",
      "size_in_bytes": 3705
    }
  ],
  "name": "vidiq-packaging-studio",
  "skill_md_contents": "---\nname: vidiq-packaging-studio\ndescription: Write, score, compare, and improve YouTube titles and thumbnails, and optionally a script, for an idea, script, reference video, or underperforming video. Use whenever a creator asks for titles or thumbnails, even a quick list; vidIQ scoring improves them. Triggers include 'title ideas', 'rewrite my title', 'thumbnail ideas', 'make or refine a thumbnail', 'which title is better', 'A/B test these titles or thumbnails', 'why didn't this video get clicks', or 'write a script for this video'.\n---\n\n# vidIQ Packaging Studio\n\nRead [Live surface notes](references/live-surface-notes.md) first, even when using only supplied evidence.\n\nCreate an honest title-thumbnail package from a concept, script, reference, or published video.\nUse the creator's supplied audience, format, language, tone, and assets.\n\n## Match the request\n\n- For a focused title or thumbnail edit, go directly to that edit. Resolve a channel or fetch\n  evidence only when the action needs it.\n- For an idea or script, establish what the content can truthfully deliver.\n- For a published video's performance diagnosis, investigate before attributing weakness to\n  packaging. Read [Research and diagnosis](references/research.md).\n- For a full package, research only gaps that can change the creative decision. Reuse sufficient\n  supplied evidence.\n- Before evaluating external references, including examples already supplied by the creator,\n  read [Discovery evidence](references/discovery-evidence.md).\n- For \"which of these directions should I use\", run the compare section below.\n\nA critique does not imply generation; a title request does not imply a thumbnail or script.\nAsk only for missing information that changes the requested result.\n\n## Develop the package\n\nPreserve what each number measures: video length does not prove how quickly an action was\ncompleted. Use a speed claim only when demonstrated or explicitly supported by the brief.\n\n1. State the viewer payoff, tension or proof, and factual limits. Preserve the supplied story and\n   numbers. When covering someone else's story, distinguish its subject from the host analyzing it.\n2. If references need to be retrieved, read [Research and diagnosis](references/research.md).\n   Transfer\n   abstract mechanisms and composition principles while using original expression. Respect the\n   creator's own or appropriately licensed assets.\n3. Develop the requested title candidates. Keep them truthful and within the platform limit.\n   Describe originality as a creative judgment; claims about other videos need supporting evidence.\n   Pair the title with thumbnail imagery or text that adds tension, proof, or stakes.\n   Judge the pair against the brief; a model score is not a performance verdict.\n4. Before using vidIQ tools to generate, score, or refine titles, thumbnails, or scripts,\n   read [Production and scoring](references/production.md).\n   Before an asynchronous call, read [Job lifecycle](references/job-lifecycle.md).\n5. Inspect completed images for factual accuracy, legibility, rights, and attribution.\n   If visual inspection is unavailable, return the draft with that limitation; do not claim it\n   passed. Keep the version that best meets the brief, explaining any score tradeoff.\n\n## Compare distinct directions\n\nWhen the creator asks which of several title and thumbnail directions to use, compare the requested\nnumber of distinct concepts (three if unspecified) for the same video. Make each a different\ntruthful promise or framing, not a cosmetic rewrite, and pair every title with thumbnail tension,\nproof, or stakes that adds information. Compare audience fit, truthful payoff, differentiation,\nproduction feasibility, and evidence; a concept can stay useful despite a lower score. Research\nreferences only when they can change the comparison, and generate images only when requested, for\nthe selected direction or the requested variants. Return the concepts with their matched titles\nand thumbnail directions, the evidence and limitations, and the recommended choice with its\nstrongest counterargument. Honor a request for comparison without forcing a winner, and never\nclaim a visual winner before inspecting renders.\n\n## Deliver\n\nReturn the requested titles, thumbnail directions, edited assets, or critique, with the rationale\nand material limitations. Include source-to-adaptation reasoning when references informed the\nwork. For executed tools, report generated URLs or pending job IDs, relevant scores and their\ncontext.\n\nOmit unused stages and empty report sections. Include a script, comparison table, or iteration log\nonly when requested or useful to understand work actually performed.\n"
}