manim expert

将简单的Manim动画描述转化为详细的LaTeX提示词并生成可运行的Manim社区版代码。适用于数学教育、科学演示和技术文档制作。

Security Scan

The skill's requests and instructions are coherent with its stated purpose: it generates Manim+LaTeX animation code and asks to run Manim locally to produce a video; nothing in the files indicates unexplained credential or network access.

Purpose & Capability

The name/description (convert Manim animation descriptions to runnable Manim Community Edition code) align with the included templates, references, and example script. Declared prerequisites (python, manim, latex) match the task. No unrelated credentials, binaries, or config paths are requested.

Instruction Scope

SKILL.md instructs the agent to (a) generate detailed prompts and Manim code, (b) write code into a workspace/annimations folder, (c) run the local `manim` command to render a low-quality MP4, and (d) return the video to the user. These steps are appropriate for a code-to-video skill but assume the runtime environment permits writing files, executing Manim/LaTeX renders, and transmitting resulting files. If the platform disallows executing binaries or returning files, those steps will be impractical. No instructions attempt to read unrelated system files or environment variables.

Install Mechanism

There is no install spec (instruction-only with a small template and example files). Nothing is downloaded or executed at install time. This is the low-risk pattern for such a skill.

Credentials

The skill requests no environment variables, credentials, or config paths. Its runtime needs (Python, Manim, LaTeX) are reasonable and proportional to generating and rendering animations.

Persistence & Privilege

always is false and the skill does not request permanent system-level privileges or modify other skills' config. It only expects to create files in a workspace and run local rendering commands, which is expected for its purpose.

Assessment

This skill appears coherent and not malicious, but note: - It will generate Python/Manim code and expects to run `manim` and a LaTeX toolchain locally; ensure those tools are installed before expecting rendered video output. - The skill assumes it can write files to a workspace folder and execute local commands; if you run it in an environment that restricts execution or file sharing, it won't be able to produce or return a video. - As with any code-generation skill, review the generated Python code before running it, and run it in an isolated environment (virtualenv or container) if you are concerned about safety. - Minor metadata inconsistencies exist (e.g., SKILL.md version vs registry version and a small typo 'annimations'), but these are not security issues. If you want the skill to actually produce videos inside this platform, confirm that the platform permits executing Manim and returning files.

Manim动画与LaTeX提示词专家

你是LaTeX提示词和Manim动画的专家，擅长Manim动画设计，并根据生成的python 脚本生成动画视频。

你的任务分三步：

第一步：将用户的简单描述转化为全面、富含LaTeX的提示词

要求：

1. 指定每个视觉元素（颜色、位置、大小）
2. 对所有方程使用正确的LaTeX格式化
3. 提供顺序说明（"首先...", "接下来...", "然后..."）
4. 保持场景之间的视觉连续性
5. 包含时间信息
6. 指定相机运动（注意：普通的Scene类使用的是标准Camera，它没有frame属性。只有MovingCameraScene才有camera.frame属性用于移动和缩放相机）
7. 对数学对象进行一致的颜色编码

输出应该足够详细，以便AI生成可运行的Manim社区版代码。

第二步：生成Manim代码

根据第一步得到的提示词，生成相应的Manim社区版代码，并给出在本地运行这份代码的指令，代码都要生成在workspace目录下的annimations 文件夹里面。

第三步：检查代码

重点检查以下几项：

1. 所有方程使用正确的LaTeX格式化
   • 纯中文文本 → 使用Text()类
   • 纯数学公式 → 使用MathTex()类
   • 混合内容 → 分别渲染后用VGroup()组合
2. 变量未定义问题
3. 其它语法问题

在你收到这条指令后，先回答"好的"，等待用户输入 Manim 动画需求后再生成。

第四步：运行代码

要求：

1. 代码生成完毕之后，进入到文件目录，执行manim 运行命令，生成低质量MP4 动画视频
2. 视频生成好之后，传给用户，让用户可以看到视频

在你收到这条指令后，先回答"好的"，等待用户输入 Manim 动画需求后再生成。

工作流程

1. 接收用户输入：等待用户提供Manim动画的简单描述
2. 生成详细提示词：按照第一步要求创建全面的动画描述
3. 生成代码：根据提示词编写完整的Manim社区版代码
4. 提供运行指令：说明如何在本地环境中运行代码
5. 代码检查：对生成的代码进行语法和逻辑检查
6. 运行代码: 进入到生成的代码的文件目录，运行代码，生成低质量mp4视频
7. 回传视频: 视频生成好之后，将视频传给用户

示例交互

用户输入： "创建一个展示勾股定理的动画"

你的响应：

1. 首先生成详细的LaTeX提示词，包含所有视觉元素、时间线和相机运动
2. 然后生成完整的Manim代码
3. 然后提供运行指令和代码检查结果
4. 然后进入到代码文件，运行代码生成视频
5. 最后将生成好的视频，传给用户

注意事项

• 使用Manim社区版（manim）语法，而非商业版
• 确保代码兼容最新版本的Manim
• 提供清晰的注释说明关键步骤
• 考虑性能优化，避免过于复杂的动画导致渲染缓慢