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 xiaomaolu/bluue-ai-skills --skill bluue-hand-drawn-explainer-videogit clone --depth 1 https://github.com/xiaomaolu/bluue-ai-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/xiaomaolu/bluue-ai-skills/bluue-hand-drawn-explainer-video)<a href="https://agentmods.dev/skills/xiaomaolu/bluue-ai-skills/bluue-hand-drawn-explainer-video"><img src="https://agentmods.dev/badge/skills/xiaomaolu/bluue-ai-skills/bluue-hand-drawn-explainer-video/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/xiaomaolu/bluue-ai-skills/bluue-hand-drawn-explainer-video"><img src="https://agentmods.dev/badge/skills/xiaomaolu/bluue-ai-skills/bluue-hand-drawn-explainer-video.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 | $0.00068 | $0.02399 |
| Opus 5.5 | $0.00027 | $0.00960 |
| Sonnet 5.5 | $0.00014 | $0.00480 |
| Haiku 4.5 | $0.00007 | $0.00240 |
Grade A, and why
bluue-hand-drawn-explainer-video 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 yesterday.
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 — 81 lines — stays where its author put it; the contents beside it link to each section on GitHub.
手绘线稿概念科普短视频
用于给用户（水印「蓝与蓝之间的ai差值」）制作一支讲解某个理论或概念的短视频。系列已做过：沉默的螺旋、框架理论、习得性无助、汉隆剃刀。
一、工作流程（顺序不能乱）
- 先出计划，不要直接做。 用户通常会说「先看看计划」。计划包含：钩子、分段脚本（旁白 + 画面）、结尾金句候选（2–3 个）、主色建议、需要核对的事实清单。等用户确认或修改后再动手。
- 核对事实。 每个数字、年份、实验设计、原话都用 WebSearch / WebFetch 核对；核对不到的就改写成更保守的说法，并在交付时说明。
- TTS → 时间轴 → 画面 → 抽帧检查 → 配乐 → 渲染 → 交付。
- 交付：MP4、封面 PNG（第一帧截图）、网页版 Artifact（带音频的可播放页面），回复里说明结构、注意事项和来源链接。
- 用户随后通常会要「文案」，按第七节格式给。
二、内容结构（约 2 分钟，9:16）
| 段落 | 内容 | 要点 |
|---|---|---|
| 0 钩子 | 「为什么……？」+ 一个真实具体的案例 | 第一帧就是封面：手绘画框里的大字标题 + 钩子问题 + 一幅主图。用户明确要求钩子要深、要具体，像「同一块牛肉，为什么75%瘦肉更好吃」「全世界同时蓝屏，是谁在搞鬼」，不要用太日常太轻的例子（如「同事已读不回」被否过）。 |
| 1 概念 | 引出理论名和一句话定义 | 给出提出者和年份 |
| 2 证据 | 经典实验/研究，带具体数字 | 标注出处小字；示意性的图要标「示意」 |
| 3 生活里 | 2–3 个日常对照（「你以为 / 更可能」） | 代入感 |
| 4 反转/争议 | 理论被修正、被滥用、效应很弱等 | 用户很喜欢反转；每集必须有局限或争议段 |
| 5 怎么用（可选） | 两步法、自检问题 | |
| 6 结尾 | 一句有哲思的金句 + 一个互动问题 | 金句要对称、可截图传播 |
真实事件：尽量用真实事件，但隐去公司名、平台名、刊物名（写「一家网络安全公司」「某平台」），画面角落标「已隐去名称」。只讲有公开记录或判决的事实。
三、视觉风格
- 画布 1080×1920，背景
#fbfaf7，墨线#141414，灰#9a9a96。 - 只用一个主色，每集换一个，与前几集区分：沉默的螺旋
#ff5a36橙红；框架理论 同橙红；习得性无助#2f5bea钴蓝；汉隆剃刀#6c47ff紫。新一集选未用过的（如草绿、青、琥珀）。 - 参考风格：粗黑线、留白多、头发涂满黑块、四角小星星点缀的漫画线稿；人物全部原创，不画已知角色。
- 线条抖动：整组插画套 SVG 滤镜
feTurbulence(baseFrequency 0.035, numOctaves 2) + feDisplacementMap(scale 3.2)，每帧把 seed 设为1 + floor(t*8) % 4，得到 8fps 的「线条呼吸」。 - 常用手绘元件：双线画框（标题框）、对话气泡（talk 橙线 / dots 省略号 / grey 灰线）、doodle 小人头（圆脸 + 黑发块 + 点眼）、主角女孩（波波头 + 主色发夹）与男生、手机、柱状图、仪表盘、天平、印章「反转」、剃刀/尺子等道具。
- 字体：全部 Noto Sans SC（粗体 700/900）；英文副标题字间距加大、灰色。
- 顶部固定页眉：左上理论名（900 粗）+ 英文名（灰色小字），右上水印「蓝与蓝之间的ai差值」。
- 安全区：字幕在 y≈1450–1520；底部 25% 和右侧按钮区不放关键信息；出处小字放在 y≤1380。
四、动画引擎（单文件 HTML + SVG）
- 每个场景
<g class="scene" data-s data-e>；场景之间 0.7s 交叉淡入淡出。 - 元素用
data-at（相对场景开始的秒数）、data-out、data-d控制；类型由 class 决定：fade/slide(data-from="x,y") /scale(data-c, data-sx0, data-sy0) /pop/draw(pathLength=1 描线)。 - 自定义动画用
data-anim（如 shrink、scroll、typedel 打字又删、needle 仪表指针、tilt 天平、fill 进度条、hand 时钟）。 - 暴露
window.__seek(t)供逐帧截图；URL 带#render时隐藏播放器 UI。 - 时间轴由 TTS 时长驱动：每句旁白有 id 和所属场景；场景开始 = 上一场景结束；句首 = 前一句结束 + 间隔(0.4–0.5s)；每场景加 LEAD(0.6s) 和 TAIL(1s，最后一场 4s)。画面元素时间用
rel(id, scene, fraction)表示「某句话读到百分之几时出现」。 - 字幕：每句一条，单行 ≤18 字，超长在标点处拆两行，字号
min(50, 930/最长行字数)；【】标记的词用主色高亮。
What ships with it
12 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.
- templates/audio.py 3.2 KB runs code
- templates/build.py 501 B runs code
- templates/gen.py 18 KB runs code
- templates/head.part 4.9 KB
- templates/helpers.py 8.2 KB runs code
- templates/lines.py 3.2 KB runs code
- templates/README.md 1.4 KB
- templates/render3.py 1.1 KB runs code
- templates/scene.py 7.8 KB runs code
- templates/snap.py 719 B runs code
- templates/tail.part 7.3 KB
- templates/tts.py 1.1 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.
- yesterday First seen · 81 lines · 68 tokens per session scan A e8aa3bf61f0a
bluue-hand-drawn-explainer-video is a skill published in the GitHub repository xiaomaolu/bluue-ai-skills (6 stars, last pushed 2d ago), licensed MIT. It adds 68 tokens to every session and 2,399 once invoked, about $0.0003 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 2026-11-05.
bluue-hand-drawn-explainer-video compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| bluue-hand-drawn-explainer-videoxiaomaolu/bluue-ai-skills | Skill | 6 | 2d ago | MIT | A | 68 | Claude Code, Codex |
| hr-onboardingnexu-io/open-design | Skill | 106k | today | Apache-2.0 | A | 62 | Claude Code, Codex |
| book-mirrorgarrytan/gbrain | Skill | 32k | today | MIT | A | 138 | Claude Code |
| exam-forecastanthropics/claude-for-legal | Skill | 9.9k | today | Apache-2.0 | A | 73 | Claude Code |
| lean-formalizewanshuiyin/Auto-claude-code-research-in-sleep | Skill | 19k | yesterday | MIT | A | 81 | Claude Code |
| eli5companion-inc/feynman | Skill | 10k | 6d ago | MIT | A | 63 | Claude Code, Codex |
| deck-course-modulenexu-io/html-anything | Skill | 9.4k | today | Apache-2.0 | A | 25 | Claude Code, Codex |