Update to PixVerse

{
  "description": "Talking head: one person speaks a script straight to camera as reusable A-roll, with a locked identity, a consistent voice, measured durations and word timing delivered for captions and graphics.",
  "included_files": [
    {
      "relative_path": "agents/openai.yaml",
      "size_in_bytes": 307
    },
    {
      "relative_path": "references/performance.md",
      "size_in_bytes": 1727
    }
  ],
  "name": "pixverse-talking-head",
  "skill_md_contents": "---\nname: pixverse-talking-head\ndescription: \"Talking head: one person speaks a script straight to camera as reusable A-roll, with a locked identity, a consistent voice, measured durations and word timing delivered for captions and graphics.\"\n---\n\n# Talking Head\n\nWhen this workflow generates images or video, apply `../../skills-shared/quality-policy.md`:\nSunburst 2K/high for images, Seedance 2.5 1080p with automatic prompt enhancement for video,\nand a stopped upgrade/fallback choice for Free/Basic or model-entitlement rejection.\n\nProduce the spoken performance itself: a person delivers exact words to camera with a stable\nidentity and voice, in takes sized to the script. The output is A-roll plus measured word\ntimes, ready for captions, boards, inserts and stickers in other workflows, or delivered as\nis. Creator ads with products use `../pixverse-ugc-video/SKILL.md`; full hosted episodes use\n`../pixverse-presenter/SKILL.md`.\n\n## Installed Command\n\nBefore running helper commands, resolve the absolute plugin root from this `./SKILL.md` path: it is two directories above this skill directory and contains `.codex-plugin/plugin.json`. In every new shell session, set:\n\n```bash\nPVX=\"<absolute-plugin-root>/scripts/pvx\"\n```\n\nRun helper commands through `\"${PVX}\"`. Do not assume the user's current working directory is the plugin checkout, and do not bypass the installed wrapper with a module invocation.\n\n## Lock Words, Person, Voice\n\nWrite or accept the script in `../../skills-shared/semantic-script.md` form: segments per\ntake, cue breaks, pronunciation for names and numbers. Measure each segment\n(`\"${PVX}\" script measure ... --rounding ceil`) and pick a supported duration with a small\nmargin; natural takes add pauses.\n\nIdentity: a supplied portrait or an accepted generated image from the phone-capture kit\n(`../../skills-shared/prompt-kits.md`), seated or standing where the brief needs, with calm\nspace where graphics will go. Voice: one short TTS reference line (`create voice`, a preset\nthat fits the person) reused on every take as an `--audios` reference. Record image, voice\nand wardrobe in project memory so later segments and variants keep them.\n\n## Generate Takes\n\nSeedance 2.5 `create reference` per segment through the gateway: host image, voice\nreference, the speaking-take kit with the segment's `dialogue` projection as the exact\nwords, one physical beat per phrase, composition stability and camera axis chosen for the\nformat. No captions, on-screen text or music in the take. Submit segments as one queue;\neach is independent, so a retake replaces only its own segment.\n\n## Measure And Deliver\n\nBuild the timeline (`../../skills-shared/word-timing.md`) as soon as takes download. Check\nalignment confidence, unmatched words and extra ASR words. Listen to flagged passages:\ncorrect the transcript if recognition is wrong; only a verified skipped or added spoken\nword calls for a retake of that segment under the effective generation policy. Deliver the takes, the timeline JSON and, when asked, captions\n(`\"${PVX}\" timeline captions ... --style clean|karaoke`) or a finished composition through\n`../../skills-shared/anchored-composition.md`. Read `./references/performance.md` for\ndelivery direction and multi-take continuity.\n\n## Execution\n\nRead `../../skills-shared/production-brief.md` for reference roles, prompt construction and revisions.\nFor media execution, follow `../../skills-shared/cli-workflow.md`; load the gateway before manual\nCLI commands or queue specs. Local media, script, timeline and graphics commands spend no credits.\nPrompt-only work ends with the usable plan and prompts; it needs no account or generation calls.\n"
}