Agent Skills › leamsigc/ShortsGenerator › ig-research-setup

ig-research-setup

GitHub

初始化社交媒体研究工具环境，检查依赖并安装缺失组件，创建项目目录与配置文件，配置Chrome调试端口以支持数据采集。

skills/ig-research-setup/SKILL.md leamsigc/ShortsGenerator

Trigger Scenarios

准备进行社交媒体数据分析需要搭建研究工具运行环境首次使用社交研究功能

Install

npx skills add leamsigc/ShortsGenerator --skill ig-research-setup -g -y

More Options

Use without installing

npx skills use leamsigc/ShortsGenerator@ig-research-setup

指定 Agent (Claude Code)

npx skills add leamsigc/ShortsGenerator --skill ig-research-setup -a claude-code -g -y

安装 repo 全部 skill

npx skills add leamsigc/ShortsGenerator --all -g -y

预览 repo 内 skill

npx skills add leamsigc/ShortsGenerator --list

SKILL.md

Frontmatter

{
    "name": "ig-research-setup",
    "version": "1.0.0",
    "description": "Setup state for social research tool. Checks prerequisites, creates directory\nstructure, creates config.json, installs deps, sets up Chrome for scraping.\nCall this before any other research state.\n",
    "allowed-tools": [
        "Bash",
        "Read",
        "Write",
        "Question"
    ]
}

Social Research — Setup State

Run this first. It checks prerequisites, creates the project, and sets up Chrome.

Steps

1. Check prerequisites

Check each of these. Ask user for permission before installing anything missing.

Homebrew (Mac only): brew --version
Node.js (v18+): node --version
Python 3: python3 --version
ffmpeg: ffmpeg -version
yt-dlp: yt-dlp --version or pip3 list 2>/dev/null | grep yt-dlp
openai-whisper: pip3 list 2>/dev/null | grep whisper

Install commands:

Homebrew: /bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"
Node (Mac): brew install node
ffmpeg (Mac): brew install ffmpeg | (Windows): winget install ffmpeg
yt-dlp: pip3 install yt-dlp
whisper: pip3 install openai-whisper

2. Create project directory

SESSION=$(date +%Y-%m-%d_%H%M%S)
mkdir -p .ig-research/projects/<project-name>/$SESSION
mkdir -p .ig-research/projects/<project-name>  # legacy root for back-compat

3. Create config.json

Ask the user:

"What's your niche? (e.g., fitness coaching, real estate, personal branding)"
"What platform? (instagram, facebook, linkedin, x, youtube)"
"What are 1-3 hashtags/keywords?"
"Any competitor accounts to analyze?"
"How many posts per keyword?"

Create .ig-research/projects/<project-name>/config.json:

{
  "name": "Project Name",
  "niche": "Niche description",
  "platform": "instagram",
  "searchTerms": ["keyword1", "keyword2"],
  "competitors": ["https://www.instagram.com/handle/"],
  "browserPort": 9222,
  "maxPostsPerSearch": 50,
  "maxCompetitorPosts": 10
}

4. Install Node dependencies

npm install --prefix "$CLAUDE_SKILL_ROOT/scripts"  # ~/.claude/skills/ig-research/scripts
# fallback: npm install --prefix ./skills/ig-research/scripts

5. Set up Chrome

Tell the user:

"Close Chrome completely (Cmd+Q on Mac, not just closing the window), then reopen it with:"

Mac: /Applications/Google\ Chrome.app/Contents/MacOS/Google\ Chrome --remote-debugging-port=9222

"Once Chrome opens, go to instagram.com and log in. Then come back and tell me when you're ready."

Wait for user confirmation before finishing.