Update to vidIQ

{
  "name": "vidiq-channel-audit",
  "description": "Audit an authorized YouTube channel over 30, 60, or 90 days. Compare performance, formats, and portfolio patterns; recommend evidence-led actions without generating content or changing state.",
  "included_files": [
    {
      "relative_path": "agents/openai.yaml",
      "size_in_bytes": 430
    },
    {
      "relative_path": "references/live-surface-notes.md",
      "size_in_bytes": 2014
    }
  ],
  "skill_md_contents": "---\nname: vidiq-channel-audit\ndescription: Audit an authorized YouTube channel over 30, 60, or 90 days. Compare performance, formats, and portfolio patterns; recommend evidence-led actions without generating content or changing state.\n---\n\n# vidIQ Channel Audit\n\nRead [Live surface notes](references/live-surface-notes.md) first, even when using only supplied evidence.\n\nProduce a private, read-only diagnosis of the authorized channel from its own evidence.\nDo not generate assets, score packaging, or make state changes.\n\n## Set comparable scope\n\nUse the requested 30-, 60-, or 90-day window; propose 30 days in the research plan if unspecified.\nChoose the latest completed calendar days, excluding today, and the immediately preceding equal\nperiod. State both exact date ranges and date basis before querying.\n\nUse a bounded analytics plan. Resolve the intended authorized channel with\n`vidiq_user_channels`; stop private analysis if it is absent instead of substituting a similarly\nnamed public channel. Request sensitive revenue, demographic, or geography breakdowns only when\nthe creator explicitly needs them.\n\n## Gather relevant reports\n\n- **Trajectory:** `vidiq_channel_analytics` with explicit dates and `dimensions: [\"day\"]`.\n  Choose supported decision-relevant metrics, such as views, watch time, average duration/\n  percentage, subscriber gains/losses, likes, and comments. Report latest observed date and\n  gaps; a returned date is not proof of complete ingestion.\n- **Format contribution:** use a separate `creatorContentType` report when the split matters.\n  Keep long-form, Shorts, and live evidence distinct and use metrics valid for that report shape.\n- **Portfolio:** a separate `video` report provides a bounded top-N view, not the full upload\n  distribution. Use `vidiq_channel_videos` for recent titles/dates only when needed, with a\n  required `videoFormat: long`, `short`, or `live` per call. Label that set a sample.\n  Filter analytics to sampled IDs using `filters: \"video==ID1,ID2\"` in batches of at most 150;\n  absent requested IDs remain unknown.\n- **Audience overlap or publishing windows:** use `vidiq_subscriber_insights` only when explicitly\n  requested for the authorized channel. Its histogram is sampled public activity,\n  not watch history or a live online-audience signal. Keep overlap data private.\n\n## Interpret valid comparisons\n\nRequire equal-day coverage and comparable formats. If shifting bounds for ingestion lag, retain\nequal full-length periods, disclose revised dates, and scope extra calls first. When an\ninside-period change matters, compare equal-day halves (15/15, 30/30, or 45/45), showing observed\nversus expected day counts. Missing values cannot be invented or unequal observed sums treated\nas equal-day performance.\n\nSum additive metrics only. Weight rates/percentages using valid available denominators;\notherwise show rows separately. Compute net subscribers as gained minus lost, report both net\ncounts and the absolute delta, and add percent change only from a positive prior net. Counts\nnever change in percentage points.\n\nAssess trajectory, watch efficiency, audience response, format contribution, and top-video\nconcentration against comparable channel history and the stated goal. Disclose cadence, upload\nages, sample size, and format-mix changes as competing explanations. Missing coverage or\nnonpositive baselines leave relative decline unknown; show available absolute deltas.\n\nDo not impose universal healthy-channel thresholds, unsupported peer benchmarks, or causal claims.\nThumbnail CTR/impressions are unavailable in this analytics schema; use supplied Studio data\nwhen needed and label its source.\n\n## Deliver\n\nGive the authorized target, periods, relevant coverage/freshness limitations, and a compact health\nsummary separating observations, interpretations, and recommendations. Include an inside-period\ncomparison only when relevant, with its coverage or availability limit.\n\nPrioritize up to three interventions, or the requested number, with evidence, competing explanation,\nconfidence, smallest action, success measure, and reassessment date. Report contradictions and\nthe next read that would most improve confidence. Route a focused launch or retention question\nto its matching skill.\n"
}