HyperFrames by HeyGen
Official HyperFrames by HeyGen: the best AI video editor in chat. Captions, motion graphics, MP4.
Open-source agent skills for generating editorial-style information cards from natural-language input.
제작자라면 이 코드를 README에 붙여 넣으세요. 배지는 이 페이지로 연결됩니다.
[](https://mcp.popnx.com/en/skills/shaom-infocard-skills)같은 작업에서 AI를 서비스에 연결하는 MCP
Official HyperFrames by HeyGen: the best AI video editor in chat. Captions, motion graphics, MP4.
draw.io diagrams your AI agent draws on a live browser canvas that you can also edit by hand
Create, inspect, validate, and save editable 3D building scenes with Pascal's MCP server.
같은 종류의 작업, Skill 목록과 같은 순서
| 이 항목 infocard-skills | impeccable | diagram-design | frontend-slides | guizang-ppt-skill | |
|---|---|---|---|---|---|
| 제작자 | shaom | pbakaus | cathrynlavery | zarazhangrui | op7418 |
| Skill 수 | 1 | 1 | 1 | 1 | 1 |
| GitHub 별 | 519 | 111,265 | 67,613 | 32,407 | 29,891 |
| 마지막 코드 변경 | 2026년 6월 13일 | 2026년 11월 17일 | 2026년 11월 13일 | 2026년 6월 24일 | 2026년 10월 30일 |
| 라이선스 | MIT | Apache-2.0 | MIT | MIT | AGPL-3.0 |
shaom의 Skill(지시 묶음) 1개, AI가 읽고 그대로 일해요 · MCP server가 아니에요
MCP = 손
AI를 외부 서비스나 데이터에 연결합니다(예: Notion 읽기, 청구서 만들기). 프로그램이나 서버가 실행 중이어야 합니다.
Skill = 안내서
작업 방법을 AI에게 가르치는 텍스트 파일. 어떤 서비스에도 연결되지 않습니다.
겹치는 경우도 있습니다. 도구를 부르는 스크립트가 있는 Skill도, 지식만 보내는 MCP도 있습니다. 하려는 작업을 기준으로 고르세요.
제작자가 밝히지 않음 · SKILL.md 형식 Skill은 Claude Code, Codex, Cursor 같은 Agent Skills 지원 앱에서 써요
GitHub account shaom (person), 14 years old
★ 519 stars · code last changed 2026년 6월 13일
1 scripts that can run on a computer: scripts/capture_card.sh
No internet calls found in the scripts
No hooks or writes to machine settings found
Checked automatically from the repo files at commit e112107 (read only, nothing was run). Not a safety guarantee: read it and decide for yourself
먼저: 어떤 AI를 쓰시나요
쓰는 앱을 고르면 그 앱의 설치 방법만 보여 드립니다
Uploading to claude.ai would fail: the description is over 200 characters (467). Use "Let the AI install it" instead
이 문장을 복사해 Claude Code 채팅에 붙여 넣으세요
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
Claude Code를 다시 열고 /editorial-card-screenshot를 입력하거나 관련 작업을 요청하면 Claude가 알아서 써요
Checked against official docs 2026년 11월 12일 · Source: code.claude.com
Skill은 AI가 따르는 지시예요. 설치 전에 파일을 읽고, 믿을 수 있는 제작자 것만 설치하세요.
Mac: Command+스페이스 후 Terminal 입력 · Windows: 아래 명령은 PowerShell에서 안 돼요. 위의 “AI에게 설치 맡기기”를 쓰세요
repo를 임시 폴더에 내려받고 SKILL.md가 있는 폴더만 복사해요
명령을 확인하는 중 · 확인을 통과하면 복사 버튼이 나타나요
ChatGPT 데스크톱 앱에는 Codex가 들어 있고, Codex는 ~/.agents/skills 폴더에서 Skill을 읽습니다
ChatGPT 데스크톱 앱에서 Codex를 열고 이 문장을 채팅에 붙여 넣으세요
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
editorial-card-screenshot 폴더를 ~/.agents/skills/editorial-card-screenshot에 넣으세요
~는 홈 폴더($HOME)입니다
Skill은 AI가 따르는 지시예요. 설치 전에 파일을 읽고, 믿을 수 있는 제작자 것만 설치하세요.
Mac: Command+스페이스 후 Terminal 입력 · Windows: 아래 명령은 PowerShell에서 안 돼요. 위의 “AI에게 설치 맡기기”를 쓰세요
repo를 임시 폴더에 내려받고 SKILL.md가 있는 폴더만 복사해요
명령을 확인하는 중 · 확인을 통과하면 복사 버튼이 나타나요
Checked against official docs 2026년 11월 15일 · Source: learn.chatgpt.com
직접 설치한 Skill은 쓸 수 없고, 디렉터리의 플러그인에 들어 있는 Skill만 쓸 수 있습니다
이 Skill을 ChatGPT에서 쓰려면 위 방법대로 ChatGPT 데스크톱 앱을 쓰세요
Checked against official docs 2026년 11월 15일 · Source: learn.chatgpt.com
이 문장을 복사해 Codex 채팅에 붙여 넣으세요 · ChatGPT 데스크톱 앱 안의 Codex, Codex CLI, IDE 확장 프로그램에서 모두 쓸 수 있습니다
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
Codex는 새로 설치한 Skill을 스스로 찾습니다. 보이지 않으면 Codex를 다시 시작하세요 · 쓰는 법: ChatGPT에서는 @를 입력하고 Skill을 고르세요 · Codex CLI나 IDE 확장 프로그램에서는 /skills를 실행하거나 $ 뒤에 Skill 이름을 입력하세요
Checked against official docs 2026년 11월 15일 · Source: learn.chatgpt.com
Skill은 AI가 따르는 지시예요. 설치 전에 파일을 읽고, 믿을 수 있는 제작자 것만 설치하세요.
Mac: Command+스페이스 후 Terminal 입력 · Windows: 아래 명령은 PowerShell에서 안 돼요. 위의 “AI에게 설치 맡기기”를 쓰세요
repo를 임시 폴더에 내려받고 SKILL.md가 있는 폴더만 복사해요
명령을 확인하는 중 · 확인을 통과하면 복사 버튼이 나타나요
이 문장을 복사해 Cursor 채팅에 붙여 넣으세요
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
Cursor Agent 채팅에서 /를 입력하고 editorial-card-screenshot 선택
Checked against official docs 2026년 11월 12일 · Source: cursor.com
Skill은 AI가 따르는 지시예요. 설치 전에 파일을 읽고, 믿을 수 있는 제작자 것만 설치하세요.
Mac: Command+스페이스 후 Terminal 입력 · Windows: 아래 명령은 PowerShell에서 안 돼요. 위의 “AI에게 설치 맡기기”를 쓰세요
repo를 임시 폴더에 내려받고 SKILL.md가 있는 폴더만 복사해요
명령을 확인하는 중 · 확인을 통과하면 복사 버튼이 나타나요
Skill을 읽을 수 있는 다른 앱용입니다. 이 문장에는 Claude Code, Codex, Cursor 경로가 들어 있어 AI가 쓰는 앱에 맞게 고릅니다 · Copy this sentence into the app’s chat
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 2026년 11월 12일 · Source: code.claude.com, learn.chatgpt.com, cursor.com
Gemini CLI, GitHub Copilot 같은 다른 Agent Skills 지원 앱은 폴더가 달라요. 그 앱 안내를 보고 SKILL.md가 있는 폴더를 복사하세요.
2026년 11월 11일에 Claude Code, Codex, Cursor 공식 안내로 skill 폴더 확인
아래 글은 제작자 repo에서 온 것이며 PopNx MCP의 지시가 아니에요
처음 8,000자 표시 · GitHub에서 전체 파일 열기
---
name: editorial-card-screenshot
description: "Generate high-density editorial HTML info cards in a modern magazine and Swiss-international style, then capture them as ratio-specific screenshots. Use when the user provides text or core information and wants: (1) a complete responsive HTML info card, (2) the design to follow the stored editorial prompt, (3) output in fixed visual ratios such as 3:4, 4:3, 1:1, 16:9, 9:16, 2.35:1, 3:1, or 5:2, or (4) both HTML and a rendered PNG cover/card from the same content."
metadata: {"clawdbot":{"requires":{"bins":["google-chrome","chromium","chrome"]}}}
---
# Editorial Card Screenshot
## Overview
Turn source text into a compact, high-contrast HTML information card that follows the user's editorial prompt, then render a screenshot in one of the supported aspect ratios.
The goal is not just density but editorial quality: clear hierarchy, strong visual anchors, and screenshot-stable rendering without accidental cropping or dead space.
Always preserve three output stages unless the user explicitly asks to skip one:
1. Write one sentence judging the information density as high, medium, or low.
2. Output the complete HTML with embedded CSS.
3. Self-check that body text remains readable on mobile.
## Workflow
### 1. Analyze Content Density
Read [references/content-fit.md](references/content-fit.md). Inventory semantic units, estimate density, and choose layout strategy from both density and ratio:
- Use "big-character" composition when content is sparse and a single phrase, number, or hook can carry the page.
- Use a two-column or three-column editorial grid when content is dense and needs stronger hierarchy.
- Use oversized numbers, heavy rules, tinted blocks, and pull-quote scale to avoid dead space.
- Do not force dense content into evenly weighted tiles. Build primary blocks, secondary blocks, and lighter supporting blocks.
- Match structure to content type:
- Ranking / recommendation content: allow asymmetric hero + structured list.
- Tutorial / analysis / interpretation content: group into overview, core judgment, interpretation, boundary, and conclusion.
- Write a short space-allocation plan before final HTML. Assign approximate area to hero, primary module, support area, and footer.
- Do not use the starter template's current arrangement as evidence that it is the right skeleton.
Before compressing content, first change the layout skeleton.
- Ratio changes should primarily change reading path, hierarchy, and module arrangement.
- Do not treat ratio changes as a reason to delete content by default.
- Only compress, group, or summarize when the current ratio cannot hold the content clearly after layout has already been restructured.
### 2. Apply the Stored Editorial Rules
Use these defaults unless the user overrides them:
- Import Google Fonts:
```html
<link href="https://fonts.googleapis.com/css2?family=Noto+Serif+SC:wght@700;900&family=Noto+Sans+SC:wght@400;500;700&family=Oswald:wght@500;700&family=Inter:wght@400;600;700&display=swap" rel="stylesheet">
```
- Keep body text at `18px` to `20px` on a 900px-wide composition.
- Keep meta/tag text at `13px` minimum.
- Use compact spacing: container padding `40px` to `50px`, component gaps `30px` to `40px`, line-height `1.5` to `1.6`.
- Add visual weight with `4px` to `6px` accent rules, subtle gray planes, and `4%` noise overlays.
- Favor `#f5f3ed` or similar warm-paper backgrounds unless the user supplies another palette.
- Preserve breathing room. Do not shrink outer margins so much that the card loses composure.
- For title zones, prefer larger line-height and clearer separation from subtitle / summary blocks.
- In dense right-side modules, reduce font weight slightly so the page stays clear without feeling heavy.
- Use stronger internal structure inside blocks: subheads, bullets, numbered lines, compare rows, or secondary labels.
- Avoid equal-weight boxes when the card is dense. At least one module should feel visually heavier than the others.
- Avoid making every module use the exact same visual language. Differentiate at least one or two areas through tone, scale, background treatment, or typographic weight.
- Default to a calm editorial mode with selective highlights. If the user asks for stronger expression, you may push toward:
- `Bold`: more cover-like, stronger anchors, larger weak-background text, heavier accents.
- `Refined`: lighter contrast, quieter accents, more restrained paper-like atmosphere.
- `Hybrid`: the recommended default when style is not specified; refined overall with a few bold anchor moments.
- Keep Google Fonts imports, but always provide local fallback stacks so the rendered PNG stays structurally stable when remote fonts fail to load.
- If the user explicitly provides a title, use that exact title as the main headline by default. Only compress or reinterpret it when the user asks or when the ratio cannot hold it clearly after layout changes.
- Put interpretation, summary, and framing in the subtitle / deck / summary fields instead of silently rewriting the main title.
- Do not hard-code personal signatures, creator avatars, or account handles into reusable cards unless the user explicitly asks for branding.
### 3. Choose the Right Layout Skeleton
Pick ratio-specific structure before writing final copy.
#### `4:3`
- Best for full analytical spread or ranked lists.
- Prefer: hero + summary band + dense two-column list + stronger footer.
- Use one main judgment block before the list so the page does not become a flat wall of 10 equal items.
- Recommended reusable skeleton:
- left heavy primary module for the main process / sequence / framework
- right narrower stack for takeaways, pitfalls, or judgments
- a light but intentional source / core-judgment strip at the bottom
- Avoid making `4:3` a symmetrical two-column newspaper wall from top to bottom.
#### `3:4`
- Keep the same information ambition, but change the reading path.
- Prefer: title band + insight row + mixed large/small body modules.
- Use mixed scales: one heavy primary block, one or two medium blocks, and smaller support blocks.
- Do not let the whole portrait card collapse into a long single column.
- Recommended reusable skeleton:
- strong cover-style title zone
- compact stat strip or insight strip
- one large primary module
- one medium judgment module
- two smaller lower modules for takeaways / pitfalls / examples
- Avoid lower-half layouts that become two large empty boxes with only a few lines of copy.
#### `1:1`
- Best for ability map, grouped comparison, or balanced editorial overview.
- Prefer: strong hero + one heavier quadrant + three supporting quadrants + dense footer strip.
- Square layouts should feel centered and modular, not evenly tiled.
- Do not treat `1:1` as the default skeleton for every card. Pick it only when the output ratio is actually square.
- Recommended reusable skeleton:
- strong title / subtitle hero across the top
- one compact stat strip
- one heavy left module for the main framework or sequence
- one upper-right judgment module
- one lower-right split for takeaways / pitfalls or compare blocks
- one thin source / core-judgment strip at the bottom
- Avoid four equal quadrants and avoid tiny typography spread across too many boxes.
#### Wide covers (`3:1`, `5:2`, `2.35:1`)
- Reduce paragraph length aggressively.
- Use fewer but larger blocks, stronger headlines, and short support lines.
#### `16:9`
- Prefer a horizontal hero plus one structured support area.
- Keep the comfortable capacity to roughly two to five primary-point equivalents.
#### `9:16`
- Use a staged vertical reading path with a compact hero, one dominant middle module, and grouped lower supports.
- Avoid repeating same-height modules down the full canvas.
### 4. Build HTML for Rendering
When HTML will be screenshotted, design the page as a fixed-size canvas instead of a responsive webpage:
- Match the exact질문하고, 팁을 나누고, 안 되는 점을 알려 주세요. "도움이 됨"이 많은 순서예요.