feat(skills): presenters go avatar first, and a voice can be stock

The person's own avatar, then Avatar IV from their photo, then Video 1 with
lipsync. With no clone or recording, a stock voice named in the credits. Pins the
five rebuilt packs. The archive test keeps its ../ under GNU tar with -P.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
Miao Yang
2026-11-02 00:52:58 +08:00
co-authored by Claude Opus 5.5
parent 6c10f1905d
commit 66a3bb78f4
4 changed files with 34 additions and 28 deletions
+1 -1
View File
@@ -106,7 +106,7 @@ describe("pullTemplate", () => {
mkdirSync(join(src, "a"));
writeFileSync(join(src, "x.md"), "x");
const out = join(dir, "evil.tar.gz");
execFileSync("tar", ["-czf", out, "-C", join(src, "a"), "../x.md"]);
execFileSync("tar", ["-czPf", out, "-C", join(src, "a"), "../x.md"]); // -P: GNU tar strips the ../ otherwise
const evil = readFileSync(out);
pin(evil);
serve({}, evil);
+21 -16
View File
@@ -17,25 +17,30 @@ export interface TemplatePack {
// ponytail: packs are pinned here, so a new pack ships with a CLI release; move to a public index if templates change faster.
export const TEMPLATES: Record<string, TemplatePack> = {
"fern-brand-explained": {
id: "M9004",
sha256: "59b2e6fc22e17a0a6133f4cf0465ecbc199fbbd2196b1c8f544b084eae67d733",
bytes: 20171637,
},
"kurzgesagt-science-explainer": {
"big-questions": {
id: "M9003",
sha256: "a508e7f1ab5fef59c2958f3e2d95deb334ea7a068b1177d585d72159c324468c",
bytes: 25563134,
sha256: "930c40664a201b5ebd04c458fcbeee3cda3d1cd7a8127a0c28750c55a8b1003c",
bytes: 25560207,
},
"vox-america-explained": {
id: "M9001",
sha256: "4d93fd4e215bbbc1751f942dfc7cbf2bb2534bcb7b78c4132e1d1fc4a2ccdc64",
bytes: 36704855,
},
"xiaolin-yikouqi": {
"long-story-short": {
id: "M9002",
sha256: "f8f8aa292a4132355cfc98ff3b62c9e0c60d0da92167b8b53538b2e87987895f",
bytes: 152166727,
sha256: "b2eef554322c6782759913a83a2885c6ea9dc8a5d710c2eebf49b8a9484ce4df",
bytes: 152171350,
},
manhunt: {
id: "M9005",
sha256: "3194bff79adf22fd645a522b812edb6147620eed499ef73cfb03850e913c24ec",
bytes: 5968498,
},
"paper-trail": {
id: "M9001",
sha256: "654e42f747f5fc31675a66ec4bbd210db6ab52aefeb7031915a2b3399ddd4944",
bytes: 36707442,
},
"under-the-hood": {
id: "M9004",
sha256: "b9aeaca877235e7d07a596d158df8c5d450780fa4b018f8ca5ce7c7bb847193b",
bytes: 20173989,
},
};
+8 -7
View File
@@ -1,24 +1,25 @@
---
name: creator-remix
description: Make a new episode in a top creator's series format - Vox-style America explainers, 小Lin 「一口气了解」, Kurzgesagt-style science films, Fern-style brand stories - on the person's topic. Use for "an episode of", "in the style of", 「用小林的模板做一集」. One video to copy is reverse-any-video.
description: Make a new episode in a proven video style - Paper Trail, Long Story Short, Big Questions, Under the Hood, Manhunt - on the person's topic. Use for "an episode of", "in the style of", 「用这个风格做一集」. One video to copy is reverse-any-video.
metadata:
internal: true
---
**Plugin installs:** Before setup or freshness commands, follow [plugin execution rules](../hyperframes/references/plugin-installation.md) when this skill is inside a HyperFrames plugin.
# A new episode from a creator's template
# A new episode from a style template
A good episode is one the series' own viewers would take for the next upload: its structure, pacing, look, graphics and sound, on a new topic, in the person's own voice and, when the series has a presenter, their own face. Never use a creator's face, voice, footage, logo or show name.
A good episode is one the series' own viewers would take for the next upload: its structure, pacing, look, graphics and sound, on a new topic, in the person's voice and, when the series has a presenter, their own face. Never use a creator's face, voice, footage, logo or show name.
Creator templates are a HyperFrames Pro feature. The CLI pulls them and the server checks the plan.
| Template | Slug | Series | Makes | The person brings |
| --- | --- | --- | --- | --- |
| Vox-style America, explained | `vox-america-explained` | `america-explained` | English explainer on one American policy or history question | Voice |
| 小Lin 一口气了解 | `xiaolin-yikouqi` | `yikouqi` | 中文主持人讲透一个国家、商品或机构 | Voice and face |
| Kurzgesagt-style science film | `kurzgesagt-science-explainer` | `science-explainer` | English fully drawn science explainer | Voice |
| Fern-style brand explained | `fern-brand-explained` | `brand-explained` | English breakdown of a brand or design, in one graphic world with real 3D | Voice |
| Paper Trail | `paper-trail` | `american-policy` | English explainer on one American policy or history question | Voice |
| Long Story Short (长话短说) | `long-story-short` | `deep-dive` | 中文主持人讲透一个国家、商品或机构 | Voice and face |
| Big Questions | `big-questions` | `science-explainer` | English fully drawn science explainer | Voice |
| Under the Hood | `under-the-hood` | `brand-explained` | English breakdown of a brand or design, in one graphic world with real 3D | Voice |
| Manhunt | `manhunt` | `crime-hunt` | English reconstruction of a hunt or a closed case, in a dark 3D world | Voice |
## Start
+4 -4
View File
@@ -21,9 +21,9 @@ Order: `read → settle the voice → say what's coming → ask (only what is op
The person's message carries what it carries. Ask once, together, only what is still open: the topic, the length, the voice and, for a presented series, the face.
- **Only the person's own voice and face.** Never a HeyGen stock voice or presenter, never the creator's face, voice, footage, wordmark or show name. A presented series with no face answered: ask for it once, and make nothing paid until it is answered.
- **A presenter runs on the plan by default.** Recommend the person's own HeyGen avatar once, in one line: it looks best, goes through the `heygen` CLI and bills that login, not the plan. It is only a recommendation. No avatar, no `heygen` or they'd rather not: say in one line that you'll make the presenter with HeyGen Video 1 on their plan, from their photo, and go on.
- **Settle the voice before any research.** A HeyGen voice id the person gave is the voice: use it and ask nothing. Anything else: with `heygen` signed in, look in `heygen voice list`. One clone: use it. Several: ask which. None, or no `heygen`: ask once for their voice id, for a clone they make in HeyGen from 30 s to 3 min of their own reading, or for their own recorded narration. Nothing else starts until this is settled.
- **Only the person's own face, and their own voice first.** Never a HeyGen stock presenter, never the creator's face, voice, footage, wordmark or show name. A presented series with no face answered: ask for it once, and make nothing paid until it is answered.
- **A presenter runs on the plan.** Their own HeyGen avatar if they have one, otherwise Avatar IV from their photo, both through `npx hyperframes media-use generate --type avatar` with the finished voice, so the lips land on it. HeyGen Video 1 is the fallback when neither works.
- **Settle the voice before any research.** A HeyGen voice id the person gave is the voice: use it and ask nothing. Anything else: with `heygen` signed in, look in `heygen voice list`. One clone: use it. Several: ask which. None, or no `heygen`: ask once for their voice id, a clone they make in HeyGen from 30 s to 3 min of their own reading, or their own recorded narration; with none of these, say in one line that you'll use a HeyGen stock voice (`npx hyperframes media-use resolve --type voice`), name it in the credits as AI-generated, and go on. Nothing else starts until this is settled.
- **Then say what's coming** in one short message that is not a question, and go on without waiting: the working title and hook, the length, and roughly how long it takes. No prices.
- **The length is the person's, and it beats the template's runtime.** Offer about 1 minute first: the cold open, the title beat and one chapter, the end card short. 2 to 4 minutes: REMIX.md's short cut. A full episode: the series' own runtime. Their own words ("90 seconds", "5 min"): that length. No length answered and none in the topic: about 1 minute, and offer the longer cut in the reply. Scale everything to it: script words (the template's script stats at that runtime), facts, sections, music cues, review rounds (one for a minute, two for a full episode); never pad or trim to reach the series' runtime instead.
- Everything else REMIX.md §0 asks (chapters, language) takes REMIX.md's default; write each one, and the length, with why, in the project's `CLAUDE.md`.
@@ -38,7 +38,7 @@ The person's message carries what it carries. Ask once, together, only what is s
- Voice: the person's voice id, through the kit's voice tool, as REMIX.md says. Then the kit's timing and pause tools. The voice is final before any presenter.
- **Audition before the whole script:** one paragraph (the kit's voice tool, as REMIX.md says), measure speech-only words a minute, and set the speed that lands on the series' rate. An instant clone takes no speed: change the words. Voice the script once; re-voice only the paragraphs that changed.
- Face (presented series): REMIX.md's host step, on the final voice: their own avatar, or HeyGen Video 1 from their photo. Their own takes: cut them per editing.md's presenter rules.
- Face (presented series): REMIX.md's host step, on the final voice: their own avatar, Avatar IV from their photo, or the HeyGen Video 1 fallback. Their own takes: cut them per editing.md's presenter rules.
- A changed line after the presenter take exists means the whole take is made again, or the lips drift.
## 6. Footage, shot plan, build