Update to vidIQ

{
  "name": "vidiq-new-upload-review",
  "description": "Assess a new YouTube upload against its channel's performance at the same publication age. Recommend whether to wait, inspect packaging, or reassess during its first hours or days.",
  "included_files": [
    {
      "relative_path": "agents/openai.yaml",
      "size_in_bytes": 418
    },
    {
      "relative_path": "references/live-surface-notes.md",
      "size_in_bytes": 2014
    }
  ],
  "skill_md_contents": "---\nname: vidiq-new-upload-review\ndescription: Assess a new YouTube upload against its channel's performance at the same publication age. Recommend whether to wait, inspect packaging, or reassess during its first hours or days.\n---\n\n# vidIQ New Upload Review\n\nRead [Live surface notes](references/live-surface-notes.md) first, even when using only supplied evidence.\n\nJudge a new upload against the channel's performance at the same publication age. Recommend\nwhether to hold, inspect, or prepare a packaging change; do not change YouTube metadata/settings.\n\n## Compare the trajectory\n\n1. Establish video, channel, goal, format, publication time, and current age from supplied context.\n   Use a bounded evidence plan.\n2. For “my channel” or private analytics, resolve the authorized target with `vidiq_user_channels`.\n   For “my newest upload,” use `vidiq_channel_videos` with `popular: false` and confirmed\n   `videoFormat: long`, `short`, or `live`, then confirm the returned video.\n3. Use `vidiq_get_videos_by_ids` for context, `vidiq_video_stats` for history, and\n   `vidiq_channel_performance_trends` for the recent-video post-publication curve.\n   Stats requires `granularity`: prefer `hourly` for early launches; use coarser intervals only\n   when age/window warrants them.\n4. Compare cumulative `views` at equivalent `minutesSincePublication`. Use returned\n   `views.perc30`, median, and `views.perc70`; missing percentiles are unavailable, not replaceable\n   by min/max. VPH is not cumulative views. Disclose age mismatch and mixed-format limitations\n   of the channel-wide curve. Do not manufacture checkpoints or extrapolate beyond observations.\n\n## Choose the next action\n\n- **Hold:** evidence is sparse, incomparable, or insufficient to justify intervention.\n- **Inspect:** repeated weakness against an age-matched baseline warrants investigation.\n  A percentile crossing alone cannot identify packaging as the cause.\n- **Prepare a change:** weakness has corroboration from supported analytics, supplied Studio\n  data, or qualitative comments, with competing explanations and confidence stated.\n  Without corroboration, keep the decision at hold or inspect.\n\nInspect only decision-relevant evidence within scope: authorized `vidiq_channel_analytics`,\nsupplied Studio CTR/impressions, or `vidiq_video_comments`. Thumbnail CTR/impressions are not in\nthe analytics metric enum; annotation CTR and ad impressions are not substitutes. Comments and\nscores cannot prove low CTR or its cause. Use `vidiq_score_title`/`vidiq_score_thumbnail` only\nwhen useful within the requested diagnosis.\n\nA known factual or formatting error can warrant correction at any age, independently of\nperformance. Choose a reassessment point from the observed channel curve, with no universal\nwaiting period or fixed checkpoint count.\n\n## Deliver\n\nReport publication age, evidence window, observed trajectory/bands and their gaps, and the\ncount/ages of checkpoints used. Give the decision, strongest evidence for and against it,\nconfidence, and next measurement/checkpoint. An unresolved video or channel prevents a trajectory\nclaim; a recommendation does not mean a change was applied.\n"
}