Agent skill

Investigation Video

by trustfuture in trustfuture/simon-skills

制作或精修不露脸商业与消费调查视频,覆盖选题、事实核查、10分钟以上叙事、真实动态素材、配音字幕、CTA,以及AI封面和各平台发布文案。按当前请求执行阶段,保留已确认稿件和品牌;不自动公开发布。

MITAuto-check passedMedia & Creative

Install Investigation Video

skills CLI
$ npx skills add trustfuture/simon-skills --skill investigation-video -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install trustfuture/simon-skills investigation-video --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/trustfuture/simon-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/investigation-video .claude/skills/investigation-video && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
investigation-video
GitHub stars
382
Token cost
~1.3k tokens
SKILL.md length
194 words
Files
44 (incl. scripts, references, assets)
Skills in repo
3
Repo updated
First seen
Licence
MIT

At a glance

制作或精修不露脸商业与消费调查视频,覆盖选题、事实核查、10分钟以上叙事、真实动态素材、配音字幕、CTA,以及AI封面和各平台发布文案。按当前请求执行阶段,保留已确认稿件和品牌;不自动公开发布。

  • Media & Creative work in your project
  • SKILL.md covers 先确定本轮范围, 按阶段推进, 项目结构 and 开场先建立观看理由, plus 6 more sections
  • Runs Python scripts from its folder

What it does

Investigation Video is an agent skill from trustfuture/simon-skills. 制作或精修不露脸商业与消费调查视频,覆盖选题、事实核查、10分钟以上叙事、真实动态素材、配音字幕、CTA,以及AI封面和各平台发布文案。按当前请求执行阶段,保留已确认稿件和品牌;不自动公开发布。

Its SKILL.md is about 1.3k tokens, which your agent loads only when the skill is triggered. The skill folder holds 46 other files, including scripts, reference files and assets (for example `README.md`, `assets/brand-stamp.py` and `assets/check-publish-copy.py`).

It sits in Media & Creative. It works with Remotion and Excalidraw. The repository describes itself as: 不露脸做视频账号的 AI skill 合集:调查长片 + 手绘白板讲解视频,一句选题出片、配音字幕、封面与各平台文案. The licence is MIT.

When your agent uses it

  • Media & Creative work in your project

Example prompts

  • “/investigation-video”

Requirements

  • Python 3

What it can do on your machine

Read from SKILL.md and the folder at commit 3ad0a25. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships 1 file in scripts/ (Python, from the files we listed), which the agent can run.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Investigation Video loads about 1.3k tokens when it runs, and up to ~15k if it reads all its reference files. Until then it costs about 29 tokens; SKILL.md has 194 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~29
When it runs · the whole SKILL.md, loaded when a task matches
~1.3k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~15k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from trustfuture/simon-skills at commit 3ad0a25, republished under its MIT licence (© trustfuture). 194 words, ~1,331 tokens.

Download SKILL.mdSave it as .claude/skills/investigation-video/SKILL.md (or your agent's skills folder). This skill also uses 43 other files; get the full folder from GitHub.
name
investigation-video
description
制作或精修不露脸商业与消费调查视频,覆盖选题、事实核查、10分钟以上叙事、真实动态素材、配音字幕、CTA,以及AI封面和各平台发布文案。按当前请求执行阶段,保留已确认稿件和品牌;不自动公开发布。

调查叙事视频

目标:故事由具体画面推进,机制由 MG 解释,结论有出处,品牌与账号配置保持一致。本包的 30 秒视觉升级 v2 样片保留为拼贴质感参考;长片同时遵循《零食称重》制作中确认的真实素材、动态叙事与发布要求。样片不代表长片素材已经够用。先看相应参考成品,再设计;参考的认可范围见各文件,不把一次认可扩大到全部后续版本。

先确定本轮范围

  • 用户要求学习对标的“画面/质感”时,围绕构图、素材合成、景深、边缘、字体、光影与动态制作展开;不要把交付转成选题、文案或叙事结构优化。
  • 延续已有选题、旁白、事实口径、画幅和已确认版本。视觉精修不等于重写稿件或更换音色;用户明确指出“没有钩子/重做开头”时,可在开头范围内改稿、重配音并检查正文衔接,保留原声线。
  • 新做完整长片时,以用户时长为准;本流程的账号目标为 10 分钟以上,可规划约 10–12 分钟,且按账号默认 1.2 倍语速(火山 speech_rate=20)实测,见素材、事实与声音。样片通常 20–30 秒;不拿样片时长当长片要求,也不靠重复画面或放慢语音凑时长。
  • 本次认可样式为横屏 16:9。其他画幅按用户和当前工程适配,不因为旧技能默认 dock 而自动增加上下品牌带、章节和进度条。
  • 用户已认可方向就继续制作。只有更换主要视觉语言时,优先做一段动态样片验证;一般视觉样片覆盖实拍、证据与MG,开场样片按钩子需要编排,不强塞三种形式;不要每次常规修改都重新索要审批。

按阶段推进

新选题、完整长片或续作先读 长片制作流程 与 管线手册。恢复当前选题、确认稿、音轨、素材、成片和发布包的版本;“继续”接着未完成项做,不重新选题或重做已确认内容。

  • 只要求选题:交付有事实入口、素材可得性和传播钩子的候选,每个候选标注核心问题类型(见叙事模式)和当下热点关联,热点题排前;不自动制作整片。
  • 完整制作:选题与事实 → 长稿与钩子 → 素材覆盖 → 配音字幕与剪辑 → CTA → 检查与交付;已有确认环节直接复用。
  • 只改视觉、CTA或发布材料:只做对应部分。发布包需求读 AI封面与发布文案;“完整交付”默认包含成片、字幕、来源与检查说明、两种AI封面和适配文案。
  • 不用普通确认点中断已经授权的制作;缺少影响事实、范围或执行的必要信息才提问。

项目结构

一期一个工程目录(projects/<date>-<slug>/:work/research、work/production、work/covers、outputs)加一个知识库目录(knowledge/<date> <题名>/材料/)。正文、来源、文案、封面、检查只在知识库维护,工程 outputs 用软链接;改镜头表后清空裁切目录重切;旧版本不删。完整树与版本规则见 项目文件夹结构。

开场先建立观看理由

写稿前先读 叙事模式与口播去套路:先定本期核心问题类型(账目、代价、规则、过程、对照、身份),连续两期不重复,开场句式黑名单不再使用。前约5秒就用具体冲突、反差或观众处境提出值得看下去的问题;不先寒暄、铺设购买情境或解释研究步骤。价格反差只是众多钩子中的一种,不是默认。钩子必须由正文中的事实支撑,不能为了吸引人暗示未经证实的供应关系、利润或违法。

开场以真实制作、拆盒、交易或事件动作推进,不用标题页、并排价格卡、大段证据截图起手。数字和短花字跟着动作出现,主体保持可见;静图加缩放、给卡片换动态背景,都不自动解决PPT感。Remotion是合成工具,不是必须采用卡片演示的视觉风格。具体编排与本次反例见长片制作流程。

品牌先行

制作前读 品牌与视觉,核对当前账号 brand 配置,并查看真实 Logo。包内示例识别是 SignPainter 手写签名+星芒+底划、强调色 #10C46F;普通粗体账号名不能替代签名字标。

  • 右上角放品牌字标,保持安静;没有左上章节、页码式标题或常驻进度条。
  • 封面字标固定:不让生图模型画签名,裁比例后用 assets/brand-stamp.py 合成,右上角、宽 = 图宽 22%、右距 4%、上距 3%,3:4 与 4:3 同规则,与成片右上角字标同几何。规则与验收见 AI封面与发布文案。
  • 采用已认可的混合层次:真实素材保留颜色;信息层可用中性浅灰、柔和光影、灰绿与信号绿。不要把整片变成暖白杂志,也不要把整片压暗调蓝。
  • 字体、色调或 Logo 的“升级”先对齐现有资产,再做细节;不要重新设计品牌。

画面与质感

视觉对标或精修先读 视觉升级 v2,查看认可成片和拼板。核心为 真实材质+前后景分离+统一拼贴处理:主体敢于放大和局部出画,背景交代环境,白描边与轻投影分离层次,少量纹理统一素材。连续动作与参数以实际工程为准,不能仅凭抽帧猜测。

真实品牌或商品优先使用官方图、实物照、原视频帧的真实抠图;生成得像真物不等于真实商品。长片中实拍与相关动态影像承担主要叙事,静图平移、粒子和文字运动不计作实拍。先完成逐段素材覆盖,再安排图解。

蛋糕、暗网格、纸条只是案例表达;按新主题选择真实物件与镜头组合,不把所有视频固定成三页模板。证据保持可读,MG 保留机制解释,沿用现有品牌和音轨。

素材与调查

读 素材、事实与声音。先从叙事提取主体、事件、订单、证照、原文和机制,再找素材:主体真容/原报道/原始证据优先,Pexels 只补少量确有用途的环境过桥。

保留来源台账、使用区间与素材原始分辨率。画面与当前句子对应;片名叫 screen 或 order 不代表内容匹配,绑定前必须查看。证据原件不能用 AI 仿造。

动画与阅读时间

读 制作与验收。在已认可素材编排上精修,不把 MG 删除后只剩背景加字。

  • MG 用于资金分配、关系、竞价等真正需要解释的信息。优先对象延续、比例拆分、路径绘制、轻微位移;所有数值和连接要有依据。
  • 短花字完整稳定停留通常至少 2.5–3 秒;短句约 4 秒起,按字数调整。
  • 证据通常安排 8–9 秒,确保入场和强调完成后仍有约 5 秒阅读;简单裁切证据可缩短,但不能只完成动画就退场。
  • 数字逐项出现后保持;最终等式至少留约 3 秒。进入/退出时间不计入阅读时间。
  • 普通素材可约 3–7 秒一切,随句义安排。快速装饰镜头不承载新文字。无需统一慢转场,优先在读完后直接切镜或按动作衔接。
  • 字幕随实际语音整句稳定显示,不逐字跳动,不附句内进度线;不能强行给每条同步字幕套 3 秒规则。

制作与交付

使用 Remotion 帧动画,音轨由 FFmpeg 合成,成片必须混入闪避配乐(配方见 管线手册)。字幕时间戳直接取火山 TTS 返回的逐字 words(2.0 音色须在 audio_params 传 enable_subtitle: true,scripts/tts.mjs 已默认打开),用 scripts/subtitles-volc.py 对齐;只有缺 words 的旧音频才回退 whisper。已有工程优先局部修改;先关键帧核对 Logo、字体、文字、素材,再整片渲染。依赖与路径查实际环境,不靠固定 localhost 端口判断当前项目。

完整成片片尾默认有简短点赞、收藏、关注CTA;结合本期价值,沿用已确认音色,重新核对新增字幕和总时长。

按本轮范围交付文件和实际检查结果;发布包中的封面默认为独立AI生图的3:4与4:3,不从视频画幅推导;字标由 assets/brand-stamp.py 后期固定合成,不让模型画。文案与封面版本须一致。样片不自动扩展成整片,制作不包含公开发布。

可复用资源

  • 公共工序 video-common(同级的 ../video-common/):事实核查与来源台账、AI 封面比例验收、平台文案机制、成片机器验收与交付边界、合规自查。本 skill 只保留本账号特有的品牌、素材、叙事与阅读时间约定。
  • 长片制作流程:选题热点栏、点名品牌的出处与申诉材料、时长、开头、素材覆盖、动态统计与CTA。
  • 叙事模式与口播去套路:六类核心问题、轮换规则、开场黑名单、写稿自检。
  • 项目文件夹结构:工程与知识库两套目录的完整树、各文件用途、版本规则。
  • 长片制作管线手册:热榜抓取、TTS/ASR 环境、证据裁图、竖屏内嵌、裁切缓存坑、配乐混音、封面与发布包的实际命令与路径。新项目从这里复制上一期工程。
  • AI封面与发布文案:3:4和4:3生图、标题、平台文案与打包。
  • 视觉升级 v2:已认可的30秒质感样例、拼板、合成做法、最小工程与复现命令。
  • 旧品牌对齐样片:早期 30 秒 MP4、拼板及机制解释参考。
  • assets/BrandStamp.jsx:可复用 Remotion/React 签名字标组件;品牌值由当前账号配置传入。
  • assets/brand-stamp.py:封面字标固定合成(右上角、22% 宽,与视频字标同几何);生图不再画签名。
  • assets/publishing-reference/八平台发布文案-模板.md + assets/check-publish-copy.py:发布文案固定模板与校验,平台顺序、字段、章节、置顶评论都固定;文案写完必须校验通过。
  • assets/finalize-outputs.py:收尾固定生成 outputs/<题名>-完整交付v<N>/(成片/字幕/使用区间实体拷贝,文档与发布包软链到知识库),每期必须用它,不手工建 outputs。
  • scripts/subtitles-volc.py:字幕对齐脚本,优先用火山逐字时间戳,缺失回退 whisper;复制到工程 work/production/ 使用。
  • assets/reference-project/:仅作已认可案例参考,不能把案例正文、日期、金额当通用默认。
  • 样片压缩版 assets/sample-preview.mp4;最小 Remotion 参考工程在 template/remotion/。

原 /shipai-changpian 仍用于主生产管线、逐字稿保护与素材审计;需要该管线时按品牌参考中的路径读取相关部分。本技能独立保留本次风格与阅读约定,原技能的旧纸张、步进条、8–10 分钟或 dock 默认不覆盖用户当前要求。

© trustfuture, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 43 other files (scripts, references, assets) in skills/investigation-video of trustfuture/simon-skills.

  • SKILL.md
  • .env.example
  • .gitignore
  • README.md
  • assets/BrandStamp.jsx
  • assets/brand-stamp.py
  • assets/check-publish-copy.py
  • assets/publishing-reference/八平台发布文案-模板.md
  • assets/sample-preview.gif
  • assets/sample-preview.jpg
  • assets/sample-preview.mp4
  • references/approved-example.md
  • references/brand-and-visual.md
  • references/covers-and-publishing.md
  • references/editorial-and-materials.md
  • references/long-form-workflow.md
  • references/narrative-modes.md
  • references/pipeline-playbook.md
  • … and 26 more

Open the folder on GitHubat commit 3ad0a25

Compare with similar skills

Investigation Video next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Investigation Video compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Investigation Video this skilltrustfuture/simon-skills382—~1.3kAutomated safety check: PassMIT
Launch Videoucsandman/marketing-studio251—~980Automated safety check: PassMIT
Content FactoryOotto-AI/claude-content-skills130—~2kAutomated safety check: PassMIT
Social Clipucsandman/marketing-studio251—~445Automated safety check: PassMIT
Remotion Best Practiceslyonjs/shortvid.io14733 repos~1kAutomated safety check: PassMIT
Canvas Capture Extensionremotion-dev/remotion63k—~907Automated safety check: PassCustom licence

Similar skills

  • Launch Video

    ucsandman/marketing-studio

    A skill your agent uses when the user wants a full launch video, hero video, or 20–60s product film combining product proof, brand, narration, music, and authored motion.

    251 GitHub stars~980 tokensUpdated 1 mo ago
    Media & CreativeAuto-check passed
  • Content Factory

    Ootto-AI/claude-content-skills

    The whole setup in one command — run the entire reel factory end to end.

    130 GitHub stars~2k tokensUpdated 2 days ago
    Media & CreativeAuto-check passed
  • Social Clip

    ucsandman/marketing-studio

    A skill your agent uses when the user wants a short social media video / feature announcement clip / X-LinkedIn-TikTok teaser for any product (e.g.

    251 GitHub stars~445 tokensUpdated 1 mo ago
    Media & CreativeAuto-check passed
  • Remotion Best Practices

    lyonjs/shortvid.io

    Best practices for Remotion - Video creation in React. An agent skill from lyonjs/shortvid.io.

    147 GitHub starsUsed in 33 repos~1k tokens
    Media & CreativeAuto-check passed
  • Canvas Capture Extension

    remotion-dev/remotion

    Official

    Rebuild, install, and reload the private Remotion Canvas Capture unpacked Chrome extension.

    63k GitHub stars~907 tokensUpdated today
    Media & CreativeAuto-check passed
  • Making Demo Videos

    nukeop/nuclear

    A skill your agent uses when making a demo, tutorial, or feature video of Nuclear.

    19k GitHub stars~892 tokensUpdated 4 days ago
    Media & CreativeAuto-check passed

More from trustfuture/simon-skills

  • Whiteboard Video

    trustfuture/simon-skills

    手绘白板风"边画边讲"讲解视频出片 skill(Excalidraw 风格逐笔动画 + 火山引擎配音 + 烧录字幕 + 品牌水印与片尾卡 + 横竖两张封面 + 各平台发布文案)。当用户说"做一期白板视频 / 边画边讲 / 手绘讲解视频 / 用 excalidraw 做视频 / whiteboard video",或要在本仓库里新建一期、改场景、换贴纸或真实…

    382 GitHub stars~1.8k tokensUpdated 16 days ago
    Auto-check: notes
  • Video Common

    trustfuture/simon-skills

    各账号视频流水线共用的四道工序(事实核查与来源台账、AI 封面比例验收、平台文案机制、成片机器验收与交付边界、合规自查)。不直接触发;由各账号 skill(如本仓库的 investigation-video-skill)在对应环节引用。音色、字幕样式、品牌、时长、口吻、平台集合都不在这里,留在各账号 skill。

    382 GitHub stars~239 tokensUpdated 16 days ago
    Auto-check passed

Questions about Investigation Video

What does Investigation Video do?

制作或精修不露脸商业与消费调查视频,覆盖选题、事实核查、10分钟以上叙事、真实动态素材、配音字幕、CTA,以及AI封面和各平台发布文案。按当前请求执行阶段,保留已确认稿件和品牌;不自动公开发布。. Investigation Video is an agent skill from trustfuture/simon-skills.

When should I use Investigation Video?

Investigation Video fits situations like: media & Creative work in your project.

How do I install Investigation Video in Claude Code?

Run `npx skills add trustfuture/simon-skills --skill investigation-video -a claude-code`. Or copy the skill folder (skills/investigation-video in trustfuture/simon-skills) into .claude/skills/investigation-video in your project. Claude Code loads it when a task matches its description.

How do I install Investigation Video in Codex?

Run `npx skills add trustfuture/simon-skills --skill investigation-video -a codex`. Or copy the skill folder (skills/investigation-video in trustfuture/simon-skills) into .agents/skills/investigation-video in your project. Codex loads it when a task matches its description.

Can I use Investigation Video in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add trustfuture/simon-skills --skill investigation-video -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/investigation-video, .gemini/skills/investigation-video, .github/skills/investigation-video and .opencode/skills/investigation-video in your project.

What does Investigation Video need to run?

Going by SKILL.md and its folder, Investigation Video needs Python for the scripts in its folder. Our summary lists: Python 3.

Does Investigation Video access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Investigation Video safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Investigation Video use?

Investigation Video is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Investigation Video use?

About 1.3k tokens (SKILL.md is roughly 5.3k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 14k tokens, read only when the agent opens those files.

What are the alternatives to Investigation Video?

Skills that share tags, products or a category with Investigation Video: Launch Video (ucsandman/marketing-studio, 251 stars), Content Factory (Ootto-AI/claude-content-skills, 130 stars), Social Clip (ucsandman/marketing-studio, 251 stars) and Remotion Best Practices (lyonjs/shortvid.io, 147 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Investigation Video?

trustfuture (a GitHub user) maintains it in trustfuture/simon-skills, which has 382 GitHub stars. The repository holds 3 skills in this directory. The repository was last updated on September 24, 2026.

Source: trustfuture/simon-skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.