{"slug":"ig-reel-extract","title":"IG Reel Extract — Download & Analyze Social Reels","tags":["instagram","reel","extract","competitor-research","transcript"],"agent_summary":"Extract Instagram Reels (and other social videos): downloads video, captures key frames, pulls transcript for competitor research and script analysis. Requires Oxylabs proxy — no home IP scraping.","trigger_phrases":["extract reel","grab this reel","instagram reel download","reel extract","ig reel","download instagram reel","reel transcript","analyze reel"],"runnable":true,"markdown":"\n# IG Reel Extract — Download & Analyze Social Reels\n\nDownload, frame-capture, and transcribe Instagram Reels for competitor research, script analysis, and content repurposing.\n\n## When to Use\n\n- Competitor research: pull scripts from top-performing competitor videos\n- Hook analysis: study what hooks work in your niche\n- Content repurposing: extract and adapt transcripts\n- Voice profiling: feed to toneprint or voice-style-extractor skill\n\n## Proxy Rule — Mandatory\n\nInstagram downloads MUST use Oxylabs proxy. No home IP scraping.\n\nRead proxy from `D:\\Ecosystem\\secrets\\MASTER_API_KEYS.env` (grep for PROXY_1).\nFormat: `host:port:user:pass` → yt-dlp proxy format: `http://user:pass@host:port`\n\n## Execution Steps\n\n### Step 1: Create output directory\n\n```bash\nmkdir D:/ClaudeDev/reel-extracts/\n```\n\nUse reel ID or timestamp as slug (e.g., `reel-2026-05-12-1830`).\n\n### Step 2: Run extractor with proxy\n\n```bash\npython \"C:\\Users\\mikem\\.claude\\skills\\ig-reel-extract\\scripts\\extract_reel.py\" \\\n \"\" \\\n --out \"D:/ClaudeDev/reel-extracts/\" \\\n --proxy \"http://user:pass@host:port\"\n```\n\n### Step 3: Read results\n\nOutput files:\n- `metadata.json` — title, author, duration, view count, like count\n- `transcript.txt` — full transcript if captions available\n- `frame_*.jpg` — key frames captured at intervals\n\n### Step 4: Analyze\n\nPresent:\n- Video downloaded: yes/no + file size\n- Frames captured: list with timestamps\n- Transcript: full text\n- Hook analysis: first 3 seconds script\n- CTA: what the video asks viewers to do\n\n## Output Folder Structure\n\n```\nD:/ClaudeDev/reel-extracts/\n reel-2026-05-12-1830/\n metadata.json\n transcript.txt\n video.mp4\n frame_0s.jpg\n frame_5s.jpg\n frame_10s.jpg\n```\n\n## Python Pattern (Direct)\n\n```python\nimport subprocess, json, os\n\ndef extract_reel(url, out_dir, proxy):\n result = subprocess.run([\n \"python\",\n r\"C:\\Users\\mikem\\.claude\\skills\\ig-reel-extract\\scripts\\extract_reel.py\",\n url,\n \"--out\", out_dir,\n \"--proxy\", proxy\n ], capture_output=True, text=True)\n \n metadata_path = os.path.join(out_dir, \"metadata.json\")\n with open(metadata_path) as f:\n return json.load(f)\n```\n\n## Hook Analysis Framework\n\nAfter extraction, analyze the transcript's first 3 seconds:\n\n| Element | Question |\n|---------|---------|\n| Pattern interrupt | Does it break expected format? |\n| Identity call-out | Does it name a specific audience? |\n| Curiosity gap | Does it create immediate question? |\n| Bold claim | Is there a specific, surprising claim? |\n| Visual hook | What happens on screen in second 1? |\n\n## Related Topics\n\n- [[social-transcript-extractor]] — extract transcripts without downloading (API method)\n- [[ytdlp-agent]] — YouTube-specific download and transcript CLI\n- [[scriptwriter]] — use extracted hooks as reference for new scripts\n- [[ai-video-pipeline]] — produce competitor-informed videos\n\n#video-sop #instagram #reel #extract #competitor-research #transcript\n","html":"

IG Reel Extract — Download & Analyze Social Reels

\n

Download, frame-capture, and transcribe Instagram Reels for competitor research, script analysis, and content repurposing.

\n

When to Use

\n
    \n
  • Competitor research: pull scripts from top-performing competitor videos
    • \n
  • Hook analysis: study what hooks work in your niche
    • \n
  • Content repurposing: extract and adapt transcripts
    • \n
  • Voice profiling: feed to toneprint or voice-style-extractor skill
    • \n
\n

Proxy Rule — Mandatory

\n

Instagram downloads MUST use Oxylabs proxy. No home IP scraping.

\n

Read proxy from D:\\Ecosystem\\secrets\\MASTER_API_KEYS.env (grep for PROXY_1).\nFormat: host:port:user:pass → yt-dlp proxy format: http://user:pass@host:port

\n

Execution Steps

\n

Step 1: Create output directory

\n
mkdir D:/ClaudeDev/reel-extracts/<slug>\n
\n

Use reel ID or timestamp as slug (e.g., reel-2026-05-12-1830).

\n

Step 2: Run extractor with proxy

\n
python \"C:\\Users\\mikem\\.claude\\skills\\ig-reel-extract\\scripts\\extract_reel.py\" \\\n  \"<URL>\" \\\n  --out \"D:/ClaudeDev/reel-extracts/<slug>\" \\\n  --proxy \"http://user:pass@host:port\"\n
\n

Step 3: Read results

\n

Output files:

\n
    \n
  • metadata.json — title, author, duration, view count, like count
    • \n
  • transcript.txt — full transcript if captions available
    • \n
  • frame_*.jpg — key frames captured at intervals
    • \n
\n

Step 4: Analyze

\n

Present:

\n
    \n
  • Video downloaded: yes/no + file size
    • \n
  • Frames captured: list with timestamps
    • \n
  • Transcript: full text
    • \n
  • Hook analysis: first 3 seconds script
    • \n
  • CTA: what the video asks viewers to do
    • \n
\n

Output Folder Structure

\n
D:/ClaudeDev/reel-extracts/\n  reel-2026-05-12-1830/\n    metadata.json\n    transcript.txt\n    video.mp4\n    frame_0s.jpg\n    frame_5s.jpg\n    frame_10s.jpg\n
\n

Python Pattern (Direct)

\n
import subprocess, json, os\n\ndef extract_reel(url, out_dir, proxy):\n    result = subprocess.run([\n        \"python\",\n        r\"C:\\Users\\mikem\\.claude\\skills\\ig-reel-extract\\scripts\\extract_reel.py\",\n        url,\n        \"--out\", out_dir,\n        \"--proxy\", proxy\n    ], capture_output=True, text=True)\n    \n    metadata_path = os.path.join(out_dir, \"metadata.json\")\n    with open(metadata_path) as f:\n        return json.load(f)\n
\n

Hook Analysis Framework

\n

After extraction, analyze the transcript's first 3 seconds:

\n

| Element | Question |\n|---------|---------|\n| Pattern interrupt | Does it break expected format? |\n| Identity call-out | Does it name a specific audience? |\n| Curiosity gap | Does it create immediate question? |\n| Bold claim | Is there a specific, surprising claim? |\n| Visual hook | What happens on screen in second 1? |

\n

Related Topics

\n
    \n
  • [[social-transcript-extractor]] — extract transcripts without downloading (API method)
    • \n
  • [[ytdlp-agent]] — YouTube-specific download and transcript CLI
    • \n
  • [[scriptwriter]] — use extracted hooks as reference for new scripts
    • \n
  • [[ai-video-pipeline]] — produce competitor-informed videos
    • \n
\n

#video-sop #instagram #reel #extract #competitor-research #transcript

\n"}