Prototype · Preview: every server in the official registry, synced every 30 minutes. Sign in with Google to write reviews.Prototype · updated every 30 min
PopNx MCPMCP servers for people who use AI · explained in plain words

auto-compare-video Skill

Repository ↗

by Cuongyd196★ 2271 skills

Install

What it does

🎬 Tạo video "so sánh kiến thức" ngắn tự động — HyperFrames + AI voice, 1 template nhiều chủ đề.

  • Connects Instruction files the AI reads
  • Account or key Not confirmed yet (README not read yet)
  • Category Coding
For makers: add a badge to your README

If you made this, paste this code into your README. The badge links back to this page.

[![Featured on PopNx MCP](https://mcp.popnx.com/badge/skill/cuongyd196-auto-compare-video.svg)](https://mcp.popnx.com/en/skills/cuongyd196-auto-compare-video)

Use together with

MCP servers that connect the AI to services for the same work

REA

Reverse engineer anything from your terminal or agent with one CLI and MCP server.

by morlutoWidely used↓ 388.7K/wk★ 466.2KUpdated 6 days ago

MCP · Connects a serviceRuns on your computer

Fallow

Codebase intelligence for TypeScript and JavaScript, exposed as structured MCP analysis tools.

by fallow-rsWidely used↓ 2.4M/wk★ 5.4KUpdated yesterday

MCP · Connects a serviceRuns on your computer

Desktop Commander

MCP server for terminal commands, file operations, and process management

by wonderwhy-erWidely used↓ 740.3K/wk★ 10.3KUpdated 6 days ago

MCP · Connects a serviceRuns on your computer

Similar Skills

Same kind of work, ordered like the Skills list

Compared with similar ones
This one
auto-compare-video
reaandrej-karpathy-skillsgraphifyMoneyPrinterTurbo
MakerCuongyd196morlutomultica-aiGraphify-Labsharry0703
Skills11111
GitHub stars227467,235234,125149,873144,332
Last code changeSep 24, 2026Nov 18, 2026Apr 20, 2026Nov 18, 2026Nov 18, 2026
License–MIT–Apache-2.0MIT

1 skills (instruction sets) by Cuongyd196 that an AI reads and follows · not an MCP server

How a Skill differs from MCP

MCP = the hands

Connects the AI to an outside service or data, such as reading Notion or creating an invoice. A program or server has to be running.

Skill = the manual

A text file that teaches the AI how to do a task. It connects to nothing.

Some overlap: a few Skills run scripts that call tools, and a few MCP servers only send knowledge. Start from the task you want done.

Works with

Claude CodeCodexCursor

From the maker’s README and the folders in the repo

Install

Is it safe?

  • Who made it

    GitHub account Cuongyd196 (person), 6 years old

  • Use and upkeep

    ★ 227 stars, few users so far · code last changed Sep 24, 2026

  • Any scripts?

    1 scripts that can run on a computer: scripts/scaffold.mjs

  • Uses the internet?

    No internet calls found in the scripts

  • Changes machine-wide settings?

    No hooks or writes to machine settings found

Checked automatically from the repo files at commit 50b1f34 (read only, nothing was run). Not a safety guarantee: read it and decide for yourself

First: which AI do you use?

Pick your app and this page shows only how to install it there

Option 1Upload to claude.ai (web and app)

  1. Download from GitHub

    This skill has no license, so we do not hand out a ZIP from our site. Download it from GitHub and arrange the folder yourself, or use "Let the AI install it" above, which is easier

    Download from GitHub (whole repo) ↗

    Unzip it, then zip only the .agents/skills/create-video folder again. The ZIP must have a create-video folder with SKILL.md at the top level, no loose files

  2. Open Skills on claude.ai and upload

    +Create skillUpload a skillpick create-video.zip

    Open claude.ai/customize/skills ↗

  3. No Upload button? Turn on file creation first

    SettingsCapabilitiesCode execution and file creation

    Turn on Code execution and file creation first (newer menus: Create and edit files) · personal plans: Settings → Capabilities · Team/Enterprise: an admin turns it on under Organization settings → Plugins & skills · every plan, including Free

Upload once; it works in claude.ai and Claude Desktop, and in recent Claude Code signed in with a Claude account (not with an API key)

Cowork and cloud sessions use the skills turned on in your claude.ai account, not ~/.claude/skills on your computer (Claude Desktop scheduled tasks run locally and still load it)

Checked against official docs Nov 12, 2026 · Source: support.claude.com, support.claude.com, code.claude.com

Option 1Let Claude Code install it

Copy this sentence and paste it into a Claude Code chat

Where to install

The AI install prompt is hidden · the destination does not match the app’s official docs · Commands hidden because they did not pass our check · install from the maker’s GitHub page instead

The AI install prompt is hidden · the destination does not match the app’s official docs · Commands hidden because they did not pass our check · install from the maker’s GitHub page instead

For one project, paste it into a chat that is open in that project’s folder · The AI always shows you the path before installing

The AI tells you what it will do first and asks your permission at every step

Restart Claude Code, then type /create-video or ask for a matching task and Claude picks it up

Checked against official docs Nov 12, 2026 · Source: code.claude.com

For Terminal users

A Skill is instructions the AI will follow. Read the files before installing, and install only from makers you trust.

  1. Open Terminal

    Mac: press Command+Space and type Terminal · Windows: the command below does not work in PowerShell; use “Let AI install it” above instead

  2. Copy the skill folder to ~/.claude/skills

    This downloads the repo to a temporary folder and copies only the folders that contain a SKILL.md

    Checking the commands · copy buttons appear once they pass

  3. Restart Claude Code, then type /create-video or ask for a matching task and Claude picks it up

Skills you install yourself work in the ChatGPT desktop app, Codex CLI and the Codex IDE extension · ChatGPT on the web and mobile cannot use skills you install yourself, only skills that come with a plugin from the directory

ChatGPT desktop app

The ChatGPT desktop app has Codex built in, and Codex reads skills from ~/.agents/skills

Option 1Let ChatGPT install it

Open Codex in the ChatGPT desktop app and paste this sentence into the chat

Where to install

The AI install prompt is hidden · the destination does not match the app’s official docs · Commands hidden because they did not pass our check · install from the maker’s GitHub page instead

The AI install prompt is hidden · the destination does not match the app’s official docs · Commands hidden because they did not pass our check · install from the maker’s GitHub page instead

The AI always shows you the path before installing

The AI tells you what it will do first and asks your permission at every step

Option 2Put the folder there yourself

Put the create-video folder at ~/.agents/skills/create-video

~ is your home folder ($HOME)

For Terminal users

A Skill is instructions the AI will follow. Read the files before installing, and install only from makers you trust.

  1. Open Terminal

    Mac: press Command+Space and type Terminal · Windows: the command below does not work in PowerShell; use “Let AI install it” above instead

  2. Copy the skill folder to ~/.agents/skills

    This downloads the repo to a temporary folder and copies only the folders that contain a SKILL.md

    Checking the commands · copy buttons appear once they pass

After installing

  • ChatGPT desktop app: open Skills in the sidebar to see that the skill is there
  • To use it: in ChatGPT type @ and pick the skill · in Codex CLI or the IDE extension run /skills or type $ and the skill name
  • Codex picks up newly installed skills by itself; if one does not show up, restart Codex

Checked against official docs Nov 15, 2026 · Source: learn.chatgpt.com

ChatGPT on the web and mobile

Skills you install yourself do not work here; only skills that come with a plugin from the directory

To use this skill with ChatGPT, use the ChatGPT desktop app as above

Checked against official docs Nov 15, 2026 · Source: learn.chatgpt.com

Option 1Let Codex install it

Copy this sentence and paste it into a Codex chat · Works in Codex inside the ChatGPT desktop app, Codex CLI and the IDE extension

Where to install

The AI install prompt is hidden · the destination does not match the app’s official docs · Commands hidden because they did not pass our check · install from the maker’s GitHub page instead

The AI install prompt is hidden · the destination does not match the app’s official docs · Commands hidden because they did not pass our check · install from the maker’s GitHub page instead

For one project, paste it into a chat that is open in that project’s folder · The AI always shows you the path before installing

The AI tells you what it will do first and asks your permission at every step

Codex picks up newly installed skills by itself; if one does not show up, restart Codex · To use it: in ChatGPT type @ and pick the skill · in Codex CLI or the IDE extension run /skills or type $ and the skill name

Checked against official docs Nov 15, 2026 · Source: learn.chatgpt.com

For Terminal users

A Skill is instructions the AI will follow. Read the files before installing, and install only from makers you trust.

  1. Open Terminal

    Mac: press Command+Space and type Terminal · Windows: the command below does not work in PowerShell; use “Let AI install it” above instead

  2. Copy the skill folder to ~/.agents/skills

    This downloads the repo to a temporary folder and copies only the folders that contain a SKILL.md

    Checking the commands · copy buttons appear once they pass

  3. Restart Codex, then ask for a matching task; Codex lists the skill

Option 1Let Cursor install it

Copy this sentence and paste it into a Cursor chat

Where to install

The AI install prompt is hidden · the destination does not match the app’s official docs · Commands hidden because they did not pass our check · install from the maker’s GitHub page instead

The AI install prompt is hidden · the destination does not match the app’s official docs · Commands hidden because they did not pass our check · install from the maker’s GitHub page instead

For one project, paste it into a chat that is open in that project’s folder · The AI always shows you the path before installing

The AI tells you what it will do first and asks your permission at every step

In Cursor’s Agent chat, type / and pick create-video

Checked against official docs Nov 12, 2026 · Source: cursor.com

For Terminal users

A Skill is instructions the AI will follow. Read the files before installing, and install only from makers you trust.

  1. Open Terminal

    Mac: press Command+Space and type Terminal · Windows: the command below does not work in PowerShell; use “Let AI install it” above instead

  2. Copy the skill folder to ~/.cursor/skills

    This downloads the repo to a temporary folder and copies only the folders that contain a SKILL.md

    Checking the commands · copy buttons appear once they pass

  3. In Cursor’s Agent chat, type / and pick create-video

Option 1Let the AI install it

For other apps that read skills. The sentence gives the paths for Claude Code, Codex and Cursor so the AI picks the one for your app · Copy this sentence into the app’s chat

Where to install

The AI install prompt is hidden · the destination does not match the app’s official docs · Commands hidden because they did not pass our check · install from the maker’s GitHub page instead

The AI install prompt is hidden · the destination does not match the app’s official docs · Commands hidden because they did not pass our check · install from the maker’s GitHub page instead

For one project, paste it into a chat that is open in that project’s folder · The AI always shows you the path before installing

The AI tells you what it will do first and asks your permission at every step

Checked against official docs Nov 12, 2026 · Source: code.claude.com, learn.chatgpt.com, cursor.com

For Terminal users

Other apps that support Agent Skills (such as Gemini CLI or GitHub Copilot) use different folders. Check that app’s guide, then copy the folder with the SKILL.md there.

View on GitHub ↗

Skill folders checked against the official Claude Code, Codex and Cursor guides on Nov 11, 2026

View the original SKILL.md

The text below comes from the maker’s repo, not from PopNx MCP

Showing the first 8,000 characters · Open the full file on GitHub

---
name: create-video
description: >
  Tạo một video MỚI cho series "so sánh / phân biệt kiến thức" của repo này — clip dọc
  TikTok/Reels/Shorts 30-40s, layout 3-zone cố định theo DESIGN.md, voiceover tiếng Việt
  sinh bằng VieNeu TTS (local), Edge TTS (miễn phí) hoặc Vbee TTS, dựng bằng HyperFrames. Dùng skill này khi người dùng nói "làm video
  so sánh X vs Y", "phân biệt X và Y", "thêm video mới vào series", "tạo video so sánh
  kiến thức", hoặc yêu cầu bất kỳ video nào theo đúng format sẵn có của repo (thư mục
  videos/<slug>/). KHÔNG dùng cho video ngoài format này (promo sản phẩm, video từ URL,
  slideshow, thêm phụ đề cho footage có sẵn).
---

# Tạo video so sánh mới

## Mục tiêu

Sinh ra một thư mục `videos/<slug>/` hoàn chỉnh, tự chạy được (`npm run check` sạch,
`npm run render` ra MP4), theo đúng layout/nhịp cố định của series: 2 card khái niệm ở nửa
trên, caption chạy từng dòng ở giữa, avatar robot MC ở nửa dưới; kịch bản 12 dòng, tổng
30-40s, voiceover TTS đo thời lượng thật để khớp animation.

Mỗi video là một project HyperFrames độc lập. **Chỉ đổi nội dung — không đổi layout.**

## Bối cảnh repo (đọc trước khi làm)

Mọi đường dẫn dưới đây tính từ **root repo** (thư mục chứa `DESIGN.md` và `videos/`):

| File | Vai trò |
|---|---|
| `DESIGN.md` | Hợp đồng layout / màu / font / motion 3-zone. **Bất biến** cho cả series. |
| `AGENTS.md` | Quy tắc chung của project HyperFrames (data-attributes, `class="clip"`, timeline paused…). |
| `vieneu.md` | Tài liệu VieNeu TTS API (CIT Voice Studio) + danh sách giọng đọc. |
| `vbee.md` | Tài liệu Vbee TTS API + danh sách `voice_code`. |
| `.env` (root, **dùng chung**) | `TTS_PROVIDER`, `VIENEU_API_URL`, `VIENEU_VOICE`, `VIENEU_SPEED`, `VBEE_APP_ID`, `VBEE_ACCESS_TOKEN`, `VBEE_VOICE_CODE`, `EDGE_VOICE`, `CHANNEL`, `AUTO_CREATE_VIDEO`. Mẫu: `.env.example`. |
| `videos/dev-vs-devops/` | **Project tham chiếu chính** — copy CSS/HTML/helper/scripts từ đây. |
| `videos/thien-thach-vs-sao-bang/` | Video đầu tiên + `BRIEF.md` bản đầy đủ. |

> ⚠️ Cả 2 video đã có đều là bản **8 dòng / 15-20s** thế hệ cũ. Chỉ copy **cấu trúc**
> (CSS, HTML skeleton, helper JS, `@font-face`) từ chúng — còn **số dòng và nhịp beat phải
> dùng bản 12 dòng / 30-40s** ở bước 1 và 4 dưới đây. Không copy y nguyên 8 dòng.

**Mọi lệnh `npm run *` chạy với cwd = `videos/<slug>/`**, không phải root repo.

## Chế độ tự động (`AUTO_CREATE_VIDEO`)

Đọc `AUTO_CREATE_VIDEO` trong `.env` ở root repo **ngay khi bắt đầu**:

- **`0` / không có (mặc định)** — giữ nguyên các điểm dừng xác nhận: chốt kịch bản (bước 1),
  hỏi lại nếu phát âm TTS sai, hỏi lại nếu tổng thời lượng lệch ngoài 30-40s, và hỏi trước
  khi render (bước 9 — tốn thời gian/tài nguyên).
- **`1`** — bỏ qua **tất cả** điểm dừng đó. Tự đề xuất và chốt luôn cặp khái niệm, kịch bản
  12 dòng, icon; tự sinh VO và tự chấp nhận phát âm (chỉ sửa phiên âm khi rõ ràng sai so với
  chính tả tiếng Việt thông thường); tự điều chỉnh nội dung nếu lệch 30-40s; chạy liên tục
  bước 1 → 9 kể cả render. Chỉ báo cáo khi xong toàn bộ, hoặc khi gặp lỗi cứng không tự sửa
  được (`npm run check` fail, TTS lỗi).

## Quy trình

### 1. Chốt nội dung trước khi code

Đề xuất phương án trước rồi để người dùng chọn/sửa (hoặc tự chốt nếu `AUTO_CREATE_VIDEO=1`):

- **Cặp khái niệm A vs B** và `slug` kebab-case (ví dụ `dev-vs-devops`, `ram-vs-rom`).
- **Góc so sánh** (1 câu) + **kịch bản đúng 12 dòng** theo nhịp chuẩn:

  | Dòng | Beat | Nội dung |
  |---|---|---|
  | 1-2 | hook | "Đây là A" / "Đây là B" |
  | 3 | nút thắt | "Sự khác nhau là gì?" |
  | 4-6 | giải A | định nghĩa → đặc điểm nổi bật → 1 ví dụ thực tế / analogy |
  | 7-9 | giải B | định nghĩa → đặc điểm nổi bật → 1 ví dụ thực tế / analogy |
  | 10-11 | so sánh trực tiếp | đối chiếu song song 2 bên (beat mới trước payoff) |
  | 12 | payoff | 1 dòng chốt, giữ khung hình tới hết — không thoát |

- **Icon 2 card** (mô tả bằng lời — sẽ vẽ CSS/SVG placeholder, **không** phụ thuộc ảnh ngoài;
  ví dụ terminal `</>` cho Dev, vòng lặp vô cực cho DevOps).

**Phát âm TTS**: nếu kịch bản có từ tiếng Anh / thuật ngữ (Dev, DevOps, AI, API, Cloud…), sinh
thử VO trước rồi hỏi người dùng Vbee đọc có đúng không. Nếu sai, sửa phiên âm tiếng Việt
**chỉ trong text đưa vào TTS** (`scripts/generate-vo.mjs`) — caption trong `index.html` vẫn giữ
chính tả gốc. Đã áp dụng: "Dev" → "Đép", "DevOps" → "Đép Ốp".

Chi tiết mẫu kịch bản + bảng phiên âm: `references/script-and-timing.md`.

### 2. Khởi tạo project

Chạy script scaffold của skill này (từ **bất kỳ** thư mục nào trong repo):

```bash
node .agents/skills/create-video/scripts/scaffold.mjs <slug>
```

Script làm hộ toàn bộ phần cơ học, dễ sai nếu làm tay:

1. `hyperframes init` với `--example blank --resolution portrait --non-interactive --skip-transcribe`
2. Gỡ thư mục con lồng trùng tên mà CLI sinh ra (`videos/<slug>/<slug>/`)
3. Xoá `CLAUDE.md` / `AGENTS.md` mà `init` sinh ra trong thư mục video (trùng với bản ở root)
4. Copy `scripts/sync-channel.mjs` + `scripts/generate-vo.mjs` từ project tham chiếu
5. Nối các npm script `sync-channel` + hook `predev` / `precheck` / `prerender` / `prepublish`
   vào `package.json`, kèm `dependencies` của video tham chiếu (`edge-tts-universal`)

Sau khi scaffold xong, cài dependencies trước khi sinh VO:

```bash
cd videos/<slug>
npm install
```

Nếu script lỗi hoặc môi trường không cho chạy, xem `references/scaffold-manual.md` để làm tay
đúng từng bước.

### 3. Sinh voiceover

Sửa `videos/<slug>/scripts/generate-vo.mjs`:

- `LINES` — 12 dòng kịch bản đã chốt (áp phiên âm TTS nếu cần, xem bước 1).
- **Giữ nguyên** cách tính `REPO_ROOT` (đọc `.env` **dùng chung ở root**) — không tạo `.env`
  riêng cho video mới.
- **Giữ nguyên** `VOICE_CODE = VBEE_VOICE_CODE || "n_hanoi_male_protrainer_education_vc"` —
  giọng đọc lấy từ `.env`, **không hardcode** giọng khác trong script.

Script hỗ trợ 2 provider, chọn bằng `TTS_PROVIDER` trong `.env` ở root:

| `TTS_PROVIDER` | Cần gì | Giọng đọc |
|---|---|---|
| `edge` (mặc định trong `.env.example`) | không cần API key — pure Node.js qua `edge-tts-universal` (dependency của video) | `EDGE_VOICE`, mặc định `vi-VN-NamMinhNeural` (nữ: `vi-VN-HoaiMyNeural`) |
| `vbee` | `VBEE_APP_ID` + `VBEE_ACCESS_TOKEN` | `VBEE_VOICE_CODE`, mặc định `n_hanoi_male_protrainer_education_vc` |

Fallback trong code là `vbee` (khi `.env` không đặt `TTS_PROVIDER`) — nhưng `.env.example` đặt
`edge` để chạy được ngay không cần tài khoản. Script in ra `TTS provider: <tên>` khi chạy — kiểm
tra dòng này nếu VO ra sai giọng.

```bash
cd videos/<slug>
node scripts/generate-vo.mjs
```

Đọc `assets/vo/durations.json` để lấy thời lượng thật từng dòng.

### 4. Tính timing

Dùng đúng công thức gap đã kiểm chứng (giữ nhịp giống bản gốc), áp cho 12 dòng:

- `start[1] = 0.55`
- Trong cùng một beat (1→2, 4→5, 5→6, 7→8, 8→9, 10→11): gap = **0.3-0.35s**
- Chuyển beat (2→3, 3→4, 6→7, 9→10, 11→12): gap = **0.4-0.45s**
- `capOut(n) = start[n] + dur[n] + 0.25` (buffer trước khi dòng thoát)
- `ROOT_DURATION ≈ start[12] + dur[12] + 1.3` (outro hold), làm tròn — **phải rơi trong
  30-40s**.

Nếu tổng chưa đạt 30s: **thêm câu / ví dụ** vào giải A/B hoặc vòng so sánh trực tiếp —
**không** kéo giãn gap để lấp thời gian (phá nhịp fast-cut của house style). Nếu vượt 40s:
cắt câu hoặc tăng `speed_rate` của Vbee.

Ví dụ tính đầy đủ: `references/script-and-timing.md`.

### 5. Dựng `index.html`

Copy cấu trúc từ `videos/dev-vs-devops/index.html`, **giữ nguyên**: biến CSS `:root`, `.card`,
`.caption-line` / `.caption-line-text`, `.kw`, avatar `#arm-left` / `#arm-right` / `#mouth`,
toàn bộ helper JS `showLine` / `pose` / `headTilt` / `talk` / glow ambient, **và khối
`@font-face` trong `<head>`**.

Chỉ đổi: `<title>`, 2 icon card, `.card-label`, 12 dòng `.caption-line`, `data-duration` của
`#root` / `#scene`, 12 thẻ `<audio>`, object `VO`, `ROOT_DURATION`, và timeline JS (nhân bản
khối `showLine` / `talk` theo nhịp 12 dòng).

Quy tắc chi tiết bắt buộc đọc trước khi viết HTML — cấu trúc caption, màu keyword, font

Discussion

Ask, share a tip or report that it does not work. Sorted by what people marked helpful.

No posts yet. Be the first to ask or share.

Posts here are written by users, not the PopNx team.

Sign in to post

From GitHub

GitHub stars
227
Maker
Cuongyd196
License
No license stated This repo states no license. By default the author keeps all rights: ask before you share it or use it commercially.
Last code change
Sep 24, 2026
SKILL.md files in the repo (including translated copies and templates)
2
Topics
aiagent, claude-code, claude-skills, codetovideo, cuongit, hyperframes, videoai