Agent Skills
› leamsigc/ShortsGenerator
› ig-research-scrape
ig-research-scrape
GitHub自动化抓取Instagram热门帖子数据，提取互动指标、标题及作者信息。支持Reels视频截图与音频下载，并整理竞品数据，输出结构化JSON结果用于社交内容研究。
Trigger Scenarios
需要获取Instagram特定标签或竞品的帖子互动数据
需要批量下载Reels视频的封面截图和音频文件
执行社交媒体内容调研任务
Install
npx skills add leamsigc/ShortsGenerator --skill ig-research-scrape -g -y
SKILL.md
Frontmatter
{
"name": "ig-research-scrape",
"version": "1.0.0",
"description": "Scrape state — runs the Instagram scraper via Chrome DevTools Protocol.\nRequires config.json to exist in the project directory.\nReads search terms + competitors from config, scrapes engagement data,\ndownloads audio, and captures hook screenshots.\n",
"allowed-tools": [
"Bash",
"Read",
"Write"
]
}
Social Research — Scrape State
Scrapes top-performing Instagram posts in a niche. Requires Chrome running with --remote-debugging-port=9222 and user logged into Instagram.
Before running
Ensure:
- Chrome is open with
--remote-debugging-port=9222flag - Instagram is logged in (tab stays open)
- Config exists at
.ig-research/projects/<project-name>/config.json
Run
node "$CLAUDE_SKILL_ROOT/scripts/scrape.js" <project-name> [sessionId]
# fallback: node ./skills/ig-research/scripts/scrape.js <project-name> --session $SESSION
Important: Keep the Instagram tab in the foreground (visible, not minimized) while this runs. Chrome throttles background tabs and screenshots will fail.
What it does
- Navigates to Instagram hashtag search pages (from config.searchTerms)
- Scrolls to collect post links
- Visits each post and extracts: likes, comments, shares, caption, author, post type
- Pauses video and captures screenshots at 0s, 1s, 2s (reels only)
- Downloads audio via yt-dlp (reels only)
- Navigates to competitor profiles (if configured) and repeats
- Deduplicates and sorts results by engagement
- Saves
raw-posts.json
Output
.ig-research/projects/<project-name>/
├── raw-posts.json # All scraped data, sorted by engagement
├── hook-screenshots/ # First 3 frames of each Reel (jpg)
└── transcripts/ # Audio files (m4a) for transcription
Platform-specific scrapers
- Instagram:
scripts/scrape.js(default) - Future:
platforms/<name>/scrape.js
Override by setting config.platform and using the corresponding scraper:
node "$CLAUDE_SKILL_ROOT/platforms/<platform>/scrape.js" <project-name> --session $SESSION
Version History
- 7b83166 Current 2026-08-29 02:31
