writing-video-prompts
Write Precise Video Generation Prompts
Video models need different motion, camera, timing, reference, and audio instructions. This skill converts settled clip concepts into focused prompts for supported model families.
使用我的智慧代理安裝
將此請求複製給你的智慧代理，其中包含權威 Skill 頁面和清單。
Review the Skillstore skill "writing-video-prompts" from https://skillstore.io/skills/supercmohq-writing-video-prompts.md and its manifest at https://skillstore.io/api/skills/supercmohq-writing-video-prompts/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.你的智慧代理仍應顯示其計畫，並依安全政策請求必要確認。
Agent 可讀資源
當 AI Agent、爬蟲或腳本需要乾淨脈絡、而不是讀取完整頁面時，請使用這些連結。
測試它
正在使用「writing-video-prompts」。 Write a five-second Veo prompt for a ceramic cup on a table. Use a slow push-in and room sound.
預期結果:
Close-up with a slow push-in. A matte ceramic cup rests on a wooden table as steam curls steadily upward in warm morning light. Ambient noise: quiet room tone and a soft ceramic tap.
正在使用「writing-video-prompts」。 Turn a two-panel skincare storyboard into an eight-second Kling prompt with one spoken line.
預期結果:
- Cut 1 (0-4s): Medium close-up. The creator lifts the closed serum bottle beside her face while the camera tracks gently forward.
- She says, "This is my morning essential." Soft room tone and a quiet fabric movement.
- Hard cut.
- Cut 2 (4-8s): Macro close-up. Her free hand tilts the intact bottle toward the light while the camera holds steady. The label remains clear.
正在使用「writing-video-prompts」。 Fix a prompt where a person holds a phone, opens a jar, and applies cream in one arm-length shot.
預期結果:
The requested framing leaves only one hand free, so the full action cannot remain physically consistent. Reframe as locked off or split the opening and application into separate clips.
安全審計
安全All 22 static findings are false positives caused by ordinary prose, Markdown code spans, file references, or scoped tool names. No shell execution, system reconnaissance, prompt injection, or malicious intent was found.
風險因素
分享與引用此報告
分享版本化評估報告、中性徽章、嵌入卡片與引用。Skillstore 只報告證據，不判斷該 Skill 是否安全。
複製報告連結
https://skillstore.io/skills/supercmohq-writing-video-prompts/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown 徽章
[](https://skillstore.io/skills/supercmohq-writing-video-prompts?utm_source=security_passport_badge)HTML 徽章
<a href="https://skillstore.io/skills/supercmohq-writing-video-prompts?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/supercmohq-writing-video-prompts/security.svg" alt="Skillstore security assessment" loading="lazy"></a>嵌入卡片
<iframe src="https://skillstore.io/embed/skills/supercmohq-writing-video-prompts.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>學術引用 (APA · BibTeX · CFF)
APA 引用
supercmohq. (2026). writing-video-prompts security audit report (audit version 1) [Author version 0.5.0]. Skillstore. https://skillstore.io/skills/supercmohq-writing-video-prompts/audits/1BibTeX 引用
@techreport{supercmohq-supercmohq-writing-video-prompts-2026,
author = {supercmohq},
title = {writing-video-prompts security audit report (audit version 1)},
institution = {Skillstore},
year = {2026},
number = {1},
url = {https://skillstore.io/skills/supercmohq-writing-video-prompts/audits/1},
note = {Author version 0.5.0}
}CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "writing-video-prompts security audit report (audit version 1)"
version: "0.5.0"
type: report
authors:
- name: "supercmohq"
date-released: "2026-08-13"
url: "https://skillstore.io/skills/supercmohq-writing-video-prompts/audits/1"
identifiers:
- type: other
value: "skillstore:supercmohq-writing-video-prompts:audit:1"
description: "Skillstore immutable audit report identifier"
Skillstore 評分
評分依據 證據置信度: 中你可以打造什麼
Direct a Product Clip
Turn product actions, camera framing, dialogue, and references into a focused prompt that preserves packaging and physical continuity.
Convert Storyboards into Cuts
Read each storyboard panel as a timed cut and produce model-specific motion, camera, performance, and audio direction.
Repair an Unstable Prompt
Simplify overloaded motion, correct timing, prevent duplicate limbs, and align references with the selected video model.
試試這些提示
Write a video prompt for [model]. The clip lasts [duration]. Show [subject] performing [single action] in [setting]. Use [framing] with [camera movement].
Write a [duration] prompt for [model] using the supplied image as the opening frame. Preserve [fixed details]. Animate [motion] with [camera behavior] and [audio].
Analyze this storyboard for a [duration] [model] clip. Create one timed cut per panel. Preserve visible positions, assign one action per cut, and include camera and audio direction.
Review this [model] prompt against the supplied media and duration. Identify timing, continuity, reference, camera, motion, and audio conflicts. Then rewrite one prompt per clip.
最佳實務
- Provide the target model, clip duration, intended action, and the role of each media asset.
- Use one clear physical action and one camera move per beat.
- Verify timed beats add up exactly and preserve visible positions between cuts.
避免
- Do not combine unrelated actions, several camera moves, and dense dialogue in one short beat.
- Do not restate every visible detail when a start frame or storyboard already fixes the composition.
- Do not ask this skill to render a finished video or guarantee model output quality.
常見問題
Does this skill generate videos?
Which video models are supported?
Can it work from a storyboard?
Can it use start frames and reference images?
Does it handle dialogue and sound?
What happens when the brief has too many actions?
開發者詳細資訊
作者
supercmohq授權
Apache-2.0
作者版本
v0.5.0
Skillstore 修訂
r1
引用
68046dfc1c93b48a04e43344644ad77d3a895012
維護新鮮度
2026/8/13
使用量
0 次下載 · 0 次瀏覽
檔案結構