smithery/hongsw

notebooklm-infographic

Use when creating professional infographics automatically using NotebookLM MCP, generating visual content from research topics with automated web search, data structuring, and browser automation for visualization

Installation

$ npx skills add smithery/hongsw --skill notebooklm-infographic

Similar popular skills

Related neighbors and high-traction skills in the same topics — useful to compare before installing.

Also in this package

Other skills from smithery/hongsw.

npx skills add smithery/hongsw

Browse all from smithery/hongsw

More details

Agent compatibility

Declared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.

Claude Code Declared
Cursor Not declared
Codex Not declared
GitHub Copilot Not declared
Windsurf Not declared
Gemini CLI Not declared
Cline Not declared
OpenCode Not declared

Skill metadata

Parsed from SKILL.md frontmatter.

Declared agents claude-code

Package contents

Files included with this skill beyond the listing page.

  • skill md SKILL.md 8,516 B
  • docs SUMMARY.md 242 B

History

  1. First recorded snapshot · 0 installs

SKILL.md

NotebookLM 인포그래픽 자동 생성

Overview

NotebookLM MCP를 활용하여 주제에 대한 전문적인 인포그래픽을 완전 자동으로 생성합니다.

Core capabilities:

  • 🔍 Automated web research and data collection
  • 📊 Data structuring and content organization
  • 🎨 Automated NotebookLM infographic generation
  • 💾 Downloadable image output
  • 🌐 Korean/English support

Total workflow time: ~5-8 minutes

When to Use

Use this skill when:

  • User requests infographic creation for any topic
  • Need to visualize research findings professionally
  • Want to combine web research with automated design
  • User invokes /infographic command or asks for visual content generation

Prerequisites

Before using this skill, ensure:

  • ✅ NotebookLM MCP installed (claude mcp add notebooklm npx notebooklm-mcp@latest)
  • ✅ Google account logged in to NotebookLM
  • ✅ Chrome browser available
  • ✅ Internet connection active
  • ✅ WebSearch MCP available
  • ✅ Browser automation tools loaded (mcpclaude-in-chrome)

Workflow Pipeline

Phase 1: Research & Data Collection (2-3 min)

Objective: Gather latest information on the topic

# Execute web searches
WebSearch(query="{TOPIC} 최신 트렌드 2026")
WebSearch(query="{TOPIC} 통계 데이터")
WebSearch(query="{TOPIC} 전망")

Extract from results:

  • Key statistics (numbers, percentages)
  • Trends (growth/decline indicators)
  • Sector classifications
  • Expert insights

Phase 2: Content Structuring (1-2 min)

Structure data using this template:

# {TOPIC}

## 핵심 전망
- **목표/지표**: [숫자]
- **성과**: [통계]
- **추가 여력**: [분석]

## 전략 원칙
1. **원칙1**: 설명
2. **원칙2**: 설명
3. **원칙3**: 설명

## 핵심 섹터/카테고리

### 1순위: [섹터명]
- 성장률: [데이터]
- 포인트: [핵심 내용]

### 2순위: [섹터명]
- 성장률: [데이터]
- 포인트: [핵심 내용]

### 3순위: [섹터명]
- 성장률: [데이터]
- 포인트: [핵심 내용]

## 실행 가이드
- Step 1: [내용]
- Step 2: [내용]
- Step 3: [내용]

## 핵심 용어
- **용어1**: 정의
- **용어2**: 정의

출처: 공개 자료 종합 정리 (날짜)

Save structured content:

Write(
    file_path="/tmp/claude-{session}/infographic_{topic}.md",
    content=structured_content
)

Phase 3: NotebookLM Notebook Creation (1-2 min)

Browser preparation:

# 1. Get browser context
tabs_context_mcp(createIfEmpty=true)

# 2. Navigate to NotebookLM
navigate(tabId=TAB_ID, url="https://notebooklm.google.com")
computer.wait(duration=2)
computer.screenshot()

Create new notebook:

# 1. Click "새로 만들기" button
computer.left_click(coordinate=[890, 100])

# 2. Enter title
computer.wait(duration=2)
computer.triple_click(coordinate=[168, 31])
computer.type(text="{TOPIC}")
computer.key(text="Return")

# 3. Verify creation
computer.screenshot()

Add source content:

# 1. Click "소스 추가" button
computer.left_click(coordinate=[184, 148])
computer.wait(duration=1)

# 2. Select "복사한 텍스트"
computer.left_click(coordinate=[761, 645])
computer.wait(duration=1)

# 3. Paste structured content
computer.left_click(coordinate=[589, 512])
computer.type(text=STRUCTURED_CONTENT)

# 4. Click "삽입" button
computer.left_click(coordinate=[758, 678])
computer.wait(duration=3)

Phase 4: Infographic Generation (1-2 min)

Trigger generation:

# 1. Find "인포그래픽" button
find(tabId=TAB_ID, query="인포그래픽 button in studio panel")

# 2. Click to generate
computer.left_click(ref="ref_142")  # Use actual ref from find result
computer.wait(duration=3)

# 3. Wait for generation
computer.wait(duration=15)
computer.screenshot()

# 4. Allow additional time if needed
computer.wait(duration=10)
computer.screenshot()

Verify result:

# 1. Locate generated infographic
find(tabId=TAB_ID, query="{generated_title}")

# 2. Open to view
computer.left_click(ref="ref_182")
computer.wait(duration=2)

# 3. Capture final result
computer.screenshot()

Phase 5: Completion Report

Report to user:

✅ 인포그래픽 생성 완료!

제목: {생성된 제목}
주요 내용:
- {섹션1}
- {섹션2}
- {섹션3}

다운로드: 화면 우측 상단 다운로드 버튼(↓) 클릭
NotebookLM 노트북: {노트북 URL}

Required Tools Reference

Tool Purpose Phase
WebSearch Information gathering 1
Write Save structured content 2
tabscontextmcp Browser preparation 3
navigate URL navigation 3
computer.screenshot Status verification 3-4
computer.left_click Button interaction 3-4
computer.type Text input 3
computer.wait Timing control 3-4
find Element location 4

Error Handling

Browser Connection Failed

try:
    navigate(url="https://notebooklm.google.com")
except:
    tabs_context_mcp(createIfEmpty=true)
    # Retry navigation

Element Not Found

# Fallback: use coordinates
computer.left_click(coordinate=[x, y])

Generation Timeout

# Wait up to 60 seconds
for i in range(4):
    computer.wait(duration=15)
    screenshot = computer.screenshot()
    if "생성 완료" in screenshot:
        break

Infographic Structure

Generated infographics typically include:

  • 핵심 전망/목표: Key goals with statistics
  • 전략/원칙: 3-5 strategic principles
  • 핵심 섹터/카테고리: Ranked categories/sectors
  • 실행 가이드: Step-by-step implementation guide

Usage Examples

Example 1: Stock Investment Guide

User: /infographic "한국 주식 투자 2026"
Claude: [Executes 5-phase workflow]
Result: Professional infographic with market outlook, strategies, key sectors

Example 2: Technology Trends

User: /ig "AI 기술 트렌드" --audience="개발자"
Claude: [Customized for developer audience]
Result: Technical infographic with AI trends and implementation paths

Example 3: Electric Vehicle Market

User: /infographic "전기차 시장" --language="en"
Claude: [English language output]
Result: EV market infographic with global statistics

Quality Checklist

Before completing, verify:

  • Topic clearly defined
  • Latest data included (2026)
  • 3-5 core sections structured
  • Visual elements rich (numbers, percentages)
  • Appropriate for target audience
  • Sources cited
  • Downloadable format available

Optimization Tips

  1. Search efficiency: Use specific keywords for targeted results
  2. Content length: 2,000-3,000 words optimal for rich infographics
  3. Structure clarity: Clear section divisions with headers
  4. Data quality: Prioritize recent statistics
  5. Visual emphasis: Highlight numbers and percentages

Common Mistakes

Mistake Fix
Vague search queries Add year, specific metrics to queries
Insufficient wait time Increase computer.wait() durations
Coordinates not working Use find() for dynamic element location
Content too short Aim for 2,000+ words with detailed sections
Missing sources Always include citation with date

Troubleshooting

"OPENAIAPIKEY not set"

export OPENAI_API_KEY=sk-your-key

NotebookLM not loading

  • Verify Google account login status
  • Check browser automation connection
  • Try manual navigation first

Infographic generation stalled

  • Increase wait duration to 30+ seconds
  • Check NotebookLM service status
  • Verify content meets minimum length requirements

Real-World Impact

This skill enables:

  • Rapid visualization: 5-8 minutes vs. hours of manual design
  • Research integration: Combines web search with visual generation
  • Professional output: NotebookLM-quality infographics
  • Scalability: Generate multiple infographics in single session

Skill Type: Technique (automation workflow) Difficulty: Intermediate (requires browser automation understanding) Dependencies: NotebookLM MCP, WebSearch MCP, Browser automation tools