parody-short
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
ChineseParody shorts
戏仿短片
A short comedy for an audience that knows the people in it, or the situation: a team's big night told as an 80s
action-show intro, a Halloween creature feature, a Christmas TV ad. The group can be any group (a sales team, a
school staff room, a five-a-side club, a family) and the subject anything they lived through together. They land harder than anything else the toolkit makes, because the
audience is laughing at things they recognise. This skill is how to make one on purpose.
Built and learned on three videos for one team in November 2026 (an action-show intro of a release, its Halloween
sister cut, a Christmas ad of A Christmas Carol). They are not in because they star real colleagues.
Then on two public shorts with an invented cast (a horror trailer about the video call that never ends, a wildlife
documentary about the return to the office), each in 16:9 and vertical, plus a brand edition in a product's own UI.
examples/这是一部面向熟悉片中人物或情境的观众的短喜剧:把团队的重要时刻讲成 80 年代动作剧片头、万圣节怪物电影,或圣诞电视广告。群体可以是任何群体(销售团队、学校教工休息室、五人制足球俱乐部、家人),主题也可以是他们共同经历过的一切。这类短片比工具包制作的其他内容都更有冲击力,因为观众是在笑自己认得出的事情。本 skill 介绍的就是如何有意识地制作此类短片。
本 skill 的构建和学习基于 2026 年 11 月为一个团队制作的三支视频(以发布为主题的动作剧片头、对应的万圣节姊妹剪辑版,以及以《圣诞颂歌》为主题的圣诞广告)。它们没有放在 中,因为片中出现的是真实同事。随后又制作了两支采用虚构演员阵容的公开短片(一支关于永远结束不了的视频通话的恐怖预告片,以及一支关于返回办公室上班的野生动物纪录片),每支都有 16:9 和竖屏两个版本,此外还有一个采用产品自身 UI 的品牌版。
examples/Why they work (in order of weight)
为什么有效(按影响权重排序)
- Real faces, moving. The step change was animating the team's profile photos. Stills got smiles; the same
people raising aviators or toasting with punch got laughs. Recognition is the engine. See .
people.md - Real in-jokes. Lines found in the team's own channels beat any written gag: a colleague's catchphrase, the
feature someone keeps cramming in, the thing someone has been saying all along. The context pass is worth more
than script polish. See .
context.md - A genre whose beats fit the event. The audience already knows the shape (three rules that get broken;
ghosts of past, present and future; the team assembling), so the video only fills the beats. The tighter the
mapping, the less the script has to explain. See .
genres.md - Permission to exaggerate. These are not release notes. Once the person asking says facts don't matter, numbers can be inflated and timelines invented. Ask once, early; then stop fact-checking jokes.
- Affection is the line. Everyone is the hero of a gag or in on it, never its victim. That keeps it warm, and it is also the safety rule (below).
- A look and sound per genre. Its own fonts, overlay and grade (VHS tracking, film grain and letterbox, candlelight), an original song or score, a few sound effects, at most one narrator line in an ad. Nothing should look like a toolkit template.
- Short. An ad is 60-90 s. A "show intro" can run 2 min. Past ~2:30 the middle sags; cut there first.
- Pick every clip. About one LTX clip in three cuts to a stranger. One wrong face breaks the spell, so contact-sheet every clip and re-roll.
- 真人的脸,会动。 真正带来质变的是让团队头像动起来。静态照片只能博得微笑;同样的人推起飞行员墨镜或举杯畅饮,则能引出笑声。辨识度是核心引擎。参见 。
people.md - 真实的内部梗。 从团队自己的频道中找到的台词,胜过任何现编的笑点:同事的口头禅、某人总要塞进去的功能,或大家一直在说的那件事。梳理语境的价值远高于打磨剧本。参见 。
context.md - 类型片的节拍与事件契合。 观众已经熟悉这种结构(三个规则被打破;过去、现在与未来的幽灵;团队集结),因此视频只需填充这些节拍。映射越严密,剧本需要解释的内容就越少。参见 。
genres.md - 允许夸张。 这不是发布说明。只要提出需求的人表示事实并不重要,数字就可以夸大,时间线也可以虚构。早期询问一次即可;之后不要再核查笑话中的事实。
- 善意是底线。 每个人都是某个笑点的主角或参与者,绝不能成为笑点的受害者。这样既能让内容保持温暖,也是必须遵守的安全规则(见下文)。
- 每种类型片都有自己的视觉与声音。 每种类型都应有专属字体、叠层和调色(VHS 跟踪噪点、胶片颗粒与遮幅黑边、烛光),原创歌曲或配乐,以及少量音效;广告中最多加入一句旁白。任何内容都不应看起来像工具包模板。
- 短。 广告为 60–90 秒。“剧集片头”可以长达 2 分钟。超过约 2 分 30 秒后,中段就会开始拖沓;届时应优先从那里删减。
- 逐个筛选每个片段。 大约每三个 LTX 片段中就有一个会剪进陌生人。一个不对的面孔就会破坏沉浸感,因此要为每个片段制作 contact sheet,并重新生成。
Two kinds of cast
两类演员阵容
- Real people (the group's own photos): the strongest laughs, for an audience that knows them. Internal by default; public only with each person's written OK.
- An invented cast (FLUX portraits, or non-human characters): for a public video, or a shared topic where the
audience recognises the situation rather than the people (video calls, the office kitchen). Recognition moves
from the faces to the situation, so the situation has to be universal and specific: "you're on mute", the
spreadsheet with 412 tabs, the jacket on the hot desk. See step 0.
people.md - Cast once, use everywhere. Generate the invented cast once in a shared folder and give each video its own costumes and clips from the same faces. Two videos with the same four people feel like a series, and a callback between them (video 2's last line is video 1's first) is a free laugh.
- 真实人物(群体成员自己的照片):笑点最强,适合认识他们的观众。默认仅供内部使用;只有获得每个人书面同意后,才可公开发布。
- 虚构演员阵容(FLUX 肖像或非人类角色):适用于公开视频,或观众因情境而非人物而认出共同话题的视频(视频通话、办公室茶水间)。此时观众的识别点从面孔转向情境,因此情境必须既普遍又具体:“你已静音”、有 412 个标签页的电子表格、搁在轮用办公桌上的外套。参见 的步骤 0。
people.md - 演员阵容一次生成,处处复用。 在共享文件夹中一次性生成虚构演员阵容,再为每支视频基于同一批面孔制作各自的服装和片段。两支视频出现同样的四个人时会让人感觉是一个系列,而在两支视频之间设置回环梗(视频 2 的最后一句呼应视频 1 的第一句)还能免费制造一个笑点。
Rules that are not optional
不可妥协的规则
- All affectionate. Nobody in the group is the butt of a joke they'd mind. No performance, health, personal life or anything said in confidence. When in doubt, cut the line.
- Real people stay internal. Their likeness belongs to them. Make it for the group, share it link-only, and ask before posting anywhere public. Anyone in it who asks to be left out is left out.
- Nothing sensitive on screen. No customer or partner names, no hostnames, no secret values (a secret is a
blurred bar or ), no internal URLs. Rename real systems ("RELAY ONE").
•••• - Original words and music. Borrow a genre's form, never a work's characters, names, lines, designs or theme
tune. A public-domain story (Dickens, Grimm, Shakespeare) can be used directly; a film or TV show cannot, renamed
or not. Write original lyrics for any song (skill).
acestep - Quote only where the audience could have read it. Public team channels, yes. DMs, no. See .
context.md
- 所有内容都要出于善意。 绝不能拿群体中的任何人取笑到他们会介意。不得涉及工作表现、健康、私人生活或任何保密内容。如有疑虑,删掉相关内容。
- 真人内容仅限内部。 真人的形象属于本人。为群体制作,仅通过链接分享;发布到任何公开渠道前先征得同意。任何要求不被拍入的人都不应出现在作品中。
- 屏幕上不得出现敏感信息。 不得出现客户或合作伙伴名称、主机名、机密值(机密内容应显示为模糊条或 )或内部 URL。将真实系统改名(如“RELAY ONE”)。
•••• - 原创文字和音乐。 只借用类型形式,绝不借用作品中的角色、名字、台词、设计或主题曲。公版故事(狄更斯、格林兄弟、莎士比亚)可以直接使用;电影或电视剧作品则不可使用,改名也不行。任何歌曲都必须创作原创歌词(skill)。
acestep - 只引用观众可能已经读过的内容。 公开团队频道中的内容可以引用,私信则不行。参见 。
context.md
Before you start
开始之前
- GPU endpoints (Modal, once): , then
uv sync --extra modal && uv run modal setup. LTX is gated: it needs a Modal secret with your Hugging Face token (uv run tools/modal_setup.py flux2 image_edit ltx2 qwen3_tts). Music: a freedocs/modal-setup.mdinACEMUSIC_API_KEY(.envskill).acestep - The project is a plain Remotion project in ; no template fits, since nothing here should look like one. Copy
projects/<name>/andpackage.jsonfrom any template (the config makesremotion.config.tsimports resolve), keep one composition per cut inlib/, and put the cast's clips insrc/Root.tsx. A shot is apublic/people/on absolute frames with the person in it:<Sequence>
tsx
import { LivingPhoto } from '../../../lib/components/LivingPhoto';
// plays the clip once and holds its last good frame; fit slows a slightly short clip to fill the shot
<Sequence from={120} durationInFrames={180}>
<LivingPhoto src="people/sam-xmas.mp4" hold={140} fit={180} pos="center 30%" style={{ inset: 0 }} />
</Sequence>- GPU 端点(Modal,只需设置一次):,然后运行
uv sync --extra modal && uv run modal setup。LTX 需要开通访问权限:必须提供一个包含你的 Hugging Face token 的 Modal secret(uv run tools/modal_setup.py flux2 image_edit ltx2 qwen3_tts)。音乐:在docs/modal-setup.md中配置免费的.env(ACEMUSIC_API_KEYskill)。acestep - 项目:这是一个位于 的普通 Remotion 项目;没有模板适合,因为这里的内容都不应看起来像模板。从任意模板复制
projects/<name>/和package.json(该配置可让remotion.config.ts导入正常解析),在lib/中为每个剪辑保留一个 composition,并将演员阵容的片段放在src/Root.tsx中。一个 shot 是基于绝对帧的public/people/,其中包含对应人物:<Sequence>
tsx
import { LivingPhoto } from '../../../lib/components/LivingPhoto';
// plays the clip once and holds its last good frame; fit slows a slightly short clip to fill the shot
<Sequence from={120} durationInFrames={180}>
<LivingPhoto src="people/sam-xmas.mp4" hold={140} fit={180} pos="center 30%" style={{ inset: 0 }} />
</Sequence>Workflow
工作流
| Step | What | Read |
|---|---|---|
| 1. Brief | Event, audience, cast (3-6 people), genre, length, "do facts matter?" | this file |
| 2. Context | Mine the group's channels and tools; write | |
| 3. Map | Genre beats onto the event; one beat per person; pick lines from the bible | |
| 4. Cast | Photos or an invented cast, costume with | |
| 5. World | FLUX backgrounds (text-free), SFX, music or a song with original lyrics | |
| 6. Build | Remotion shots on absolute frames; | |
| 7. Check | Contact sheet of the render, | |
| 8. Cuts | Vertical, and a brand edition if wanted: new layouts over the same assets | |
Shots, not scenes. These are edited like ads: a fixed shot list with absolute frame ranges, a 10-20 frame
dissolve between shots, the song or score under everything. If there is narration, generate it first and anchor
to it (CLAUDE.md, Audio-Anchored Timelines); an ad with one closing narrator line can be music-led.
Sound on a movement: measure, don't eyeball. A sneeze, a jacket landing, a badge jingle: place each sound
with (where its peak meets the movement), list it in the project's ,
and run on every render before showing it. A hand-placed sneeze was half a
second early; three other cues were 0.8-1.0 s off. If a clip freezes or cuts, end its sound there too.
tools/sync_check.py --clipsync-cues.jsonsync_check.py --video ... --cuesText is the punchline carrier. Speech bubbles, a stamp, a dance card, a tombstone, a typed label. Remotion
renders all of it (FLUX and LTX garble text). Give every on-screen gag a frame cue and a sound.
Voices. For a narrator nobody owns, design one with Qwen VoiceDesign from a written brief ("a warm,
middle-aged British woman narrating a wildlife documentary, hushed, a slow measured pace"; "a deep, gravelly movie
trailer announcer"): . VoiceDesign makes a new voice on every call, so design once, then clone that reference
for every sentence (three takes, ). Describe a genre, never a real person; picking a narrator the
genre's famous voice isn't (a woman for a wildlife film) keeps it clearly original.
qwen3_tts.py --design-instruct "..." --text "<a 20 s seed paragraph in the genre, not the script>" --format wavpick_takes.pyCuts. Plan the vertical cut and any brand edition from the start; they reuse every generated asset and cost
only a render each. See .
formats.mdReuse across genres. A cast animated once can be re-costumed for the next holiday: the photos, crops and
context bible carry over, and only the costumes, clips and world change.
| 步骤 | 内容 | 参考 |
|---|---|---|
| 1. 需求概述 | 事件、受众、演员阵容(3–6 人)、类型、时长、“事实重要吗?” | 本文件 |
| 2. 语境梳理 | 挖掘该群体各频道和工具中的信息;编写 | |
| 3. 类型映射 | 将类型片节拍映射到事件;每人分配一个节拍;从 bible 中挑选台词 | |
| 4. 演员阵容 | 使用照片或虚构演员阵容;通过 | |
| 5. 世界构建 | FLUX 背景(不含文字)、SFX,以及音乐或带原创歌词的歌曲 | |
| 6. 制作 | 基于绝对帧搭建 Remotion 镜头;用 | |
| 7. 检查 | 渲染结果的 contact sheet、 | |
| 8. 剪辑版本 | 竖屏版,以及按需制作的品牌版:在相同素材上使用新布局 | |
用镜头,而非场景。 剪辑方式应像广告:使用带绝对帧范围的固定镜头表,镜头之间加入 10–20 帧的叠化,并在所有画面下方铺歌曲或配乐。如果有旁白,应先生成旁白并以其作为时间线锚点(CLAUDE.md、Audio-Anchored Timelines);只有一句收尾旁白的广告可以由音乐主导。
动作音效:对齐要靠测量,不能凭感觉。 打喷嚏、外套落下、工牌叮当作响:用 放置每段声音,使其峰值与动作发生的时刻吻合;将它们列入项目的 ,并在每次展示前对每次渲染运行 。有一次手动放置的喷嚏音效早了半秒;另外三个提示点则偏差了 0.8–1.0 秒。如果片段冻结或被剪断,声音也要在相应位置结束。
tools/sync_check.py --clipsync-cues.jsonsync_check.py --video ... --cues文字是笑点的载体。 对话气泡、印章、舞会卡、墓碑、键入的标签。Remotion 可以渲染所有这些内容(FLUX 和 LTX 会把文字弄乱)。为每个屏幕笑点设置帧提示点和音效。
旁白。 如果旁白不沿用任何人的声音,请根据书面 brief 使用 Qwen VoiceDesign 设计一个(“一位温暖、适中年龄的英国女性,以轻声、缓慢而克制的节奏旁白一部野生动物纪录片”;“一位低沉、沙哑的电影预告片播音员”):。VoiceDesign 每次调用都会生成一个新声音,因此应先设计一次,再以该参考声音克隆出每一句台词(拍摄三条,使用 )。描述一种类型风格,绝不要描述真人;不要选用与该类型片著名声音相同的旁白(例如为野生动物纪录片选择女性旁白),这样才能确保声音明显属于原创。
qwen3_tts.py --design-instruct "..." --text "<a 20 s seed paragraph in the genre, not the script>" --format wavpick_takes.py剪辑版本。 从一开始就规划竖屏剪辑和品牌版;它们会复用每一项生成的素材,只需各渲染一次。参见 。
formats.md跨类型复用。 演员阵容制作一次动画后,就可以在下一个节日重新换装:照片、裁剪结果和 context bible 都可以沿用,只有服装、片段和世界需要改变。
Cost (Modal, 2026-11)
成本(Modal,2026-11)
LTX ~$0.08 a 5 s clip, image_edit ~$0.01 an edit warm (a cold start is ~10 min; send one edit first and wait),
FLUX ~$0.01 a background or portrait, ACE-Step song a few cents. A 90 s ad with five cast members and re-rolls is
about $2. Two 80-100 s shorts sharing one invented cast (9 costume edits, 16 clips, 2 designed voices) came to about
$2.50 together. The acemusic API can return 502s mid-batch; a failed take is skipped, so re-run for more.
LTX 每个 5 秒片段约 $0.08, 在暖启动状态下每次编辑约 $0.01(冷启动约需 10 分钟;先提交一次编辑并等待),FLUX 每张背景或肖像约 $0.01,一首 ACE-Step 歌曲只需几分钱。一支 90 秒广告包含五名演员阵容成员以及重新生成的片段,总计约 $2。两支共用同一虚构演员阵容的 80–100 秒短片(9 次换装编辑、16 个片段、2 个设计声音)合计约 $2.50。acemusic API 在批量处理中途可能返回 502;失败的 take 会被跳过,因此需要重新运行以获得更多结果。
image_editRelated
相关内容
context.mdgenres.mdpeople.mdformats.mdltx2qwen-editacestepelevenlabsvideo-designremotionlib/components/LivingPhoto.tsx本目录中的 、、、;、、、、、 等 skill;以及 。
context.mdgenres.mdpeople.mdformats.mdltx2qwen-editacestepelevenlabsvideo-designremotionlib/components/LivingPhoto.tsx