Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add aliang2052/aliang-skills --skill aliang-shortvideogit clone --depth 1 https://github.com/aliang2052/aliang-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/aliang2052/aliang-skills/aliang-shortvideo)<a href="https://agentmods.dev/skills/aliang2052/aliang-skills/aliang-shortvideo"><img src="https://agentmods.dev/badge/skills/aliang2052/aliang-skills/aliang-shortvideo/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/aliang2052/aliang-skills/aliang-shortvideo"><img src="https://agentmods.dev/badge/skills/aliang2052/aliang-skills/aliang-shortvideo.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00242 | $0.02571 |
| Opus 5 | $0.00121 | $0.01286 |
| Sonnet 5 | $0.00048 | $0.00514 |
| Haiku 4.5 | $0.00024 | $0.00257 |
Grade A, and why
aliang-shortvideo scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 108 lines — stays where its author put it; the contents beside it link to each section on GitHub.
短剧全流程制作流水线
把一句话灵感做成可播短剧,五阶段、每步暂停确认:
① 剧情大纲 → ② 分集剧本(台词) → ③ 分镜表(枢纽) → ④ 批量出图 → ⑤ 图生视频(成片)
⚠️ 视频生成费钱。①~④ 可放心跑;⑤ 图生视频默认不自动实跑——开发/测试/演示时只产出每段的 bl video 命令交给用户,或等用户显式确认要为某段付费,再逐段生成。功能要做全,钱不替用户烧。
前置
bl(百炼 CLI)可用并已登录(bl auth status可查)。出图阶段需要它。- 开工先确认 4 件事,其余用默认:① 一句话灵感(必问);② 题材(用 AskUserQuestion 给 3-4 个高频项让用户选,见
references/genres.md);③ 总时长(用 AskUserQuestion 给 15s / 30s / 60s + 自定义;用户已在灵感里说了时长就别再问);④ 画风(用 AskUserQuestion 从references/styles.md的 22 个预设里让用户挑,分写实/动漫/3D-CG/插画/风格化 5 类;按题材把最搭的放第一标「(推荐)」。问画风时务必明确邀请上传参考图:在选项里加一项「📎 我发一张参考图(照着这张的画风做)」,并提示用户可直接把图片拖/粘贴进对话、或给本地路径/URL。用户发了图就走参考图模式,见下条)。画风是画面质量的头号变量,必须开工就定,别拖到出图。比例默认竖版 9:16、单镜默认 5s,用户没要求就别追问。 - 时长换算:图片张数 ≈ 总时长 ÷ 5(每 5s 一镜一张)。15s≈3 张、30s≈6 张、60s≈12 张。总时长定了,镜数就定了。
- 可选·参考图:用户甩了参考图(海报/截图/角色图)时,别只靠肉眼翻译——① 用
bl vision把它读成一段精准的style串;② 把图路径填进storyboard.json的style_ref,出图脚本会用bl image edit直接迁移这张图的画风,还原度更高。两者可并用。
工作流
阶段 ① 剧情大纲
- 确认题材后,读
references/genres.md取该题材的爽点库 + 结构模板 + 红线。 - 抽爽点:跑
python3 scripts/pick_beats.py --genre <题材>(真随机抽 8-12 个,避免套路雷同)。 - 按该题材结构写约 500 字纯叙事大纲,融入抽到的爽点,遵守反套路 + 内容健康红线。
- 输出大纲 + 列出本次用的爽点 + 逆袭/反转路径标注。
- 暂停:请用户确认大纲走向,改完再进下一阶段。
阶段 ② 分集剧本
把大纲扩成可拍剧本:
- 按节奏切场(每场一个地点+冲突单元),标注 场景 / 人物 / 动作 / 台词。
- 台词口语化、短、有冲突感;旁白克制。
- 黄金 3 秒原则:第一场就抛冲突/悬念。
- 短剧分集可选:若用户要多集,每集结尾留扣。
- 暂停:用户确认剧本,再进分镜。
阶段 ③ 分镜表(枢纽)
读 references/storyboard.md,产出两份同源数据:
- 给人看的分镜表格(镜号/时长/景别/运镜/画面/角色动作/台词/情绪)。
- 给脚本用的
storyboard.json(存到项目目录),含全片统一style、固定characters[].desc、每镜image_prompt。
关键:
- 镜数由总时长定:镜数 ≈ 总时长 ÷ 5(单镜默认 5s)。先按这个铺满镜数,再按剧情节奏微调。
- 单镜时长用户可逐镜调:分镜表每镜都有「时长」列,用户可指定某镜 5s/10s/15s。给某镜 10s/15s 时,按 5s 为单位拆成多个子镜(视频阶段每段 5s),图片阶段每个子镜出 1 张——所以加长某镜会增加该镜的图片数。拆完把总时长/总张数报给用户对一下。
image_prompt只写画面,不写台词/镜号。- 落实画风:把开工选定的画风预设(
references/styles.md里的 style 串)写进storyboard.json的style;有参考图就填style_ref。此处只做应用/微调,不重新问(开工已定)。 - 先定全片
style串和主角desc(一致性命脉,见 storyboard.md 第六节)。 - 暂停:用户确认分镜 + 各镜时长 + 画风,再出图。
What ships with it
6 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 108 lines · 242 tokens per session scan A f51c63f5b9f5
aliang-shortvideo is a skill published in the GitHub repository aliang2052/aliang-skills (26 stars, last pushed 3mo ago), licensed MIT. It adds 242 tokens to every session and 2,571 once invoked, about $0.0012 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
dramaclaw
A skill for answering identity and self-introduction questions, and for working with the DramaClaw/NovelVideo pipeline. The description also covers turning novels or stories into short vertical videos.
ai-comic-script
A Chinese-language guide for creating complete scripts for AI-generated animated short dramas, from the initial idea through episode planning and finished scenes.
ai-comic-visual-assets
A skill that turns a script into reusable AI image prompts for its main characters, settings, and props. It also defines file names, folders, and reference-image steps for visual consistency.
ai-comic-storyboard
A storyboard tool for turning an existing AI-comic script into a shot-by-shot production table. A storyboard lists what appears in each shot, how the camera moves, what characters do, the dialogue or sound, the timing, and the assets used.
ai-comic-video-prompts
A skill that converts an existing shot list—a plan describing each video shot—into one complete prompt per shot for AI video tools such as Vidu. It includes the shot framing, camera movement, actions, dialogue, sound effects, and referenced assets.
comic-mv-director
A production planner for animated stories, short dramas, music videos, and other AI video projects. It first decides whether the project is music-led or story-led, then plans storyboards, images, video, music, and editing.