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 ChenShuo2004/cs-skills --skill cs-knowledge-filmgit clone --depth 1 https://github.com/ChenShuo2004/cs-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/chenshuo2004/cs-skills/cs-knowledge-film)<a href="https://agentmods.dev/skills/chenshuo2004/cs-skills/cs-knowledge-film"><img src="https://agentmods.dev/badge/skills/chenshuo2004/cs-skills/cs-knowledge-film/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/chenshuo2004/cs-skills/cs-knowledge-film"><img src="https://agentmods.dev/badge/skills/chenshuo2004/cs-skills/cs-knowledge-film.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 | $0.00105 | $0.01807 |
| Opus 5.5 | $0.00042 | $0.00723 |
| Sonnet 5.5 | $0.00021 | $0.00361 |
| Haiku 4.5 | $0.00011 | $0.00181 |
Grade A, and why
cs-knowledge-film 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 18d 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 — 90 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CS 暗夜知识片
把一个抽象概念讲成一部安静、克制、有电影感的知识短片。本 skill 的暗夜星空视觉是用户选择这一风格时使用的实现，不是所有知识视频的默认视觉。先找到选题的核心问题与可见的动作，再决定使用现有场景、扩展场景，或为该题材另做画面。
交付
- 一个项目目录：
spec.json（唯一需要手写的文件）、audio/、timeline.json、subtitles.srt、preview.jpg、video.mp4。 - 视频默认 1920×1080、30fps、H.264 + AAC；可选中文配音（edge-tts）+
score.py生成的配乐与音效（参照原片：A 小调低音长音、温暖和弦铺底、玻璃钟琴、与画面事件逐一对齐的打字/转轮/呼啸/撞击声，-16 LUFS）；画面内置中英双语字幕，同时导出subtitles.srt。 - 不上传、不发布。BGM 只用用户提供的文件（
spec.bgm）。
风格速查（细节见 references/style.md）
- 底色近黑带一点藏蓝，慢速闪烁的星空，暗角 + 轻微颗粒。
- 字体：中文宋体/衬线（Noto Serif SC），英文 Lora 斜体；不用黑体。
- 颜色只有三种语义：金＝重点/惊奇/答案，冰蓝＝结构/线条/已知，绯红＝噪声/错误/被排除。
- 左上角章节标签
01 猜 THE GUESS；底部中文字幕 + 英文斜体小字。 - 动效慢：淡入 0.6–1s、缓出曲线；光点有辉光和呼吸感；不要弹跳、不要转场特效。
工作流
1. 先定三件事
缺了才问，能自己判断就不问：
- 主题与一句话结论（这支片子最后让观众记住哪一句）
- 时长（默认 2–3 分钟；每分钟约 220–250 字口播）
- 是否需要英文字幕（默认要，和参考片一致）
2. 写口播稿（见 references/script.md）
先提炼一句话结论、主要误解和适合这个题目的画面隐喻，再设计钩子与章节。可参考“体验式钩子 → 标题 → 3–5 章 → 回到开头收束”，但故事结构由内容决定，不必重复参考片的章节或结尾。
每句 8–25 字，一句一个意思，适合被逐句打在字幕上。数字和史实要核对，不编造。
3. 拆场景写 spec.json（见 references/spec.md）
把 assets/example-spec.json 当作格式示例，不复制其中的场景顺序或视觉隐喻。先列出每章要让观众看见的变化，再选现有场景类型；现有类型表达不了时，按 references/spec.md 的扩展说明新增场景。
| 概念形态 | 场景类型 |
|---|---|
| 预期 / 猜下一个字 / 填空 | typewriter |
| 候选词概率、采样、温度 | probs |
| 片名 | title |
| 二分、度量、量化 | numberline |
| 罕见事件、时间、希望、灯火 | landscape（烽火、时间条、弧线、灯） |
| 日常 vs 罕见新闻 | city（夜景、下雪、头条） |
| 公式、定义 | formula |
| 语言、冗余、噪声、压缩 | textplay |
| 传输、编码、信道 | signal / probe |
| 同样数据不同意义 | compare |
| 信息茧房、同质化 | cards |
| 名人原话 | quote |
| 真实图片（星图、照片） | image |
| 停顿、提问 | orb / blank |
事件时间用 "L2"（本场第 2 句开始）、"L2e"（第 2 句结束）、"L2+0.5" 锚定到口播，配音变长变短都不会错位。
4. 配音 → 时间轴 → 预览 → 渲染
把 scripts/ 下 4 个文件放在任意位置，项目目录里只放 spec.json：
python scripts/tts.py <项目目录> # 逐句配音 audio/line_XXX.mp3（需联网，免费）
node scripts/timeline.mjs <项目目录> # 按真实音频长度排时间轴，拼 narration.wav + subtitles.srt
python scripts/score.py <项目目录> # 按场景和事件生成配乐 + 音效 score.wav（改了 spec 就重跑）
node scripts/render.mjs <项目目录> --preview # 每场 2 帧拼成 preview.jpg —— 必须先看
node scripts/render.mjs <项目目录> --still 42 # 单帧排查
node scripts/render.mjs <项目目录> # 全量渲染 video.mp4
node scripts/render.mjs <项目目录> --remux # 只换声音：复用已渲染画面重新混音
What ships with it
13 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.
- agents/openai.yaml 423 B
- assets/example-spec.json 8.4 KB
- assets/examples/ai-hallucination.json 6.2 KB
- assets/examples/ai-next-word.json 9.2 KB
- references/qa.md 1.4 KB
- references/script.md 1.6 KB
- references/spec.md 7.5 KB
- references/style.md 2.1 KB
- scripts/engine.html 47 KB
- scripts/render.mjs 7.9 KB runs code
- scripts/score.py 17 KB runs code
- scripts/timeline.mjs 3.4 KB runs code
- scripts/tts.py 2.4 KB runs code
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.
- 18d ago Changed · -5 lines ca375522cfdd
- 23d ago First seen · 95 lines · 105 tokens per session scan A 4a122afe7c23
cs-knowledge-film is a skill published in the GitHub repository ChenShuo2004/cs-skills (204 stars, last pushed 18d ago), licensed MIT. It adds 105 tokens to every session and 1,807 once invoked, about $0.0004 per session on Opus 5.5. 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 2026-10-02.
Other skills, from other repositories
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chronicle
Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
component-fixtures
Use when creating or updating component fixtures for screenshot testing or their shared infrastructure, or when designing UI components to be fixture-friendly. Covers fixture file structure, theming, service setup, CSS scoping, async rendering, validation, and common pitfalls.