🤖 Tutorial: Video Downloading, Thread Unrolling & Media Export with Claude

Home › Docs › Claude Tutorials › Tutorial: Video Downloading, Thread Unrolling & Media Export with Claude
Claude Tutorials

Tutorial: Video Downloading, Thread Unrolling & Media Export with Claude

You are my X/Twitter media specialist. I want to use XActions to download videos, unroll threads into readable content, export media, and build a content archive. Help me save the best content from X.

Context

I'm using XActions (https://github.com/nirholas/XActions), an open-source X/Twitter toolkit with video downloading (x_download_video, src/scrapers/videoDownloader.js), thread unrolling (scripts/threadUnroller.js, src/scrapers/threadUnroller.js), media scraping, and content export.

What I Need You To Do

Part 1: Download Videos from Tweets

  1. Via MCP using x_download_video:

    "Download the video from this tweet: https://x.com/user/status/123456"

    Parameters:

    • tweetUrl: URL of the tweet containing video

    Returns:

    • Direct video download URLs (multiple quality options when available)
    • Video resolution and format info
    • Thumbnail URL

  2. How the video downloader works:

    • Navigates to the tweet page via Puppeteer
    • Extracts video URLs from the page's React state / network requests
    • X stores videos on their CDN (video.twimg.com)
    • Returns direct URLs you can open in a browser or download with wget/curl

  3. Download the actual file:

    # Using the URL returned
curl -o video.mp4 "VIDEO_URL_HERE"

# Or with wget
wget -O video.mp4 "VIDEO_URL_HERE"

  4. Via browser script (scripts/videoDownloader.js):

    • Navigate to the tweet with the video
    • Paste the script in DevTools
    • It extracts video URLs from the page
    • Shows download links in the console

  5. Via CLI:

    # If implemented in CLI
xactions download-video "https://x.com/user/status/123456"

Part 2: Thread Unrolling

Save threads as clean, readable content:

  1. Via MCP — Use x_get_tweets to pull the thread:

    "Unroll this thread: https://x.com/user/status/123456"
"Get all tweets in this thread and format as readable text"

  2. Via browser script (scripts/threadUnroller.js or src/scrapers/threadUnroller.js):

    • Navigate to the first tweet of the thread
    • Paste the script
    • It automatically:
      • Detects the thread structure
      • Follows the reply chain
      • Extracts all tweet text in order
      • Includes media references
      • Formats for readability

  3. Output formats:

    • Plain text: Clean readable text
    • Markdown: Headers, links, formatting preserved
    • JSON: Structured data with metadata

  4. Via CLI:

    # Unroll to terminal
xactions thread "https://x.com/user/status/123456"

# Save as markdown
xactions thread "https://x.com/user/status/123456" --format markdown --output thread.md

# Save as JSON with metadata
xactions thread "https://x.com/user/status/123456" --format json --output thread.json

  5. What's captured per tweet in a thread:

    • Tweet text (full content)
    • Author and username
    • Timestamp
    • Engagement stats (likes, retweets, replies)
    • Media URLs (images, videos)
    • Tweet position in thread (1 of N)

Part 3: Media Scraping

Export all media from any profile:

  1. Via CLI:

    xactions media USERNAME --limit 100 --format json --output media.json

  2. Via browser script (scripts/scrapeMedia.js):

    • Navigate to the user's Media tab (x.com/USERNAME/media)
    • Paste the script
    • Scrolls through all media
    • Extracts image and video URLs
    • Exports as JSON

  3. What you get:

    • Image URLs (multiple resolutions available on X's CDN)
    • Video URLs with quality options
    • Original tweet text and URL
    • Timestamp

  4. Bulk download media:

    # Export URLs then download
xactions media photographer1 --format json | jq -r '.[].urls[]' > urls.txt

# Download all with wget
wget -i urls.txt -P downloads/

# Or with parallel downloading
cat urls.txt | xargs -P 4 -I {} wget {} -P downloads/

Part 4: Bookmark Export & Content Library

Build a personal content library from X:

  1. Export bookmarks:

    "Export all my bookmarks to JSON"

    or

    xactions bookmarks --limit 500 --format json --output my-bookmarks.json

  2. Export likes:

    # Via browser script scripts/scrapeLikes.js
# Navigate to your Likes page, paste the script

  3. Organize exported content:

    • Threads → Markdown files in a knowledge base
    • Videos → Downloaded locally
    • Images → Saved for reference/inspiration
    • Tweets → Categorized by topic

Part 5: Viral Tweet Discovery & Download

Find and archive the best content:

  1. Find viral content:

    "Search for viral tweets about 'AI' with minimum 1000 likes"

  2. Via scripts/viralTweetsScraper.js or src/scrapers/viralTweets.js:

    • Search by keyword or from specific accounts
    • Filter by minimum engagement
    • Export top-performing tweets

  3. Build a swipe file:

    • Archive viral tweets in your niche
    • Analyze what makes them work
    • Use as inspiration (never copy!)
    • Track patterns: hooks, formats, topics

Part 6: Link Scraping

Extract all links shared by any user:

  1. Via browser script (src/automation/linkScraper.js or scripts/linkScraper.js):

    • Navigate to a user's profile
    • Paste core.js, then linkScraper.js
    • Extracts all URLs from their tweets
    • Groups by domain (which websites they share most)
    • Exports the link list

  2. Use cases:

    • Find what blogs/tools someone recommends
    • Map someone's go-to resources
    • Discover hidden gems in your niche
    • Competitive intelligence (what links do competitors share?)

Part 7: Reply & Quote Tweet Scraping

  1. Scrape replies (scripts/scrapeReplies.js):

    • Navigate to any tweet
    • Paste the script
    • Extracts all replies with author info and engagement
    • Great for sentiment analysis on specific tweets

  2. Scrape quote tweets (scripts/scrapeQuoteRetweets.js):

    • Navigate to a tweet
    • Paste the script
    • Extracts all quote tweets (people's reactions and commentary)
    • Useful for seeing how your content spreads

Part 8: Notification Export

Save your notification history:

  1. Via MCP:

    "Export my last 200 notifications"

  2. Via browser script (scripts/scrapeNotifications.js):

    • Navigate to Notifications tab
    • Paste the script
    • Extracts and categorizes all notifications:
      • Likes
      • Retweets
      • Follows
      • Mentions
      • Replies

Part 9: Spaces Content

Save content from X Spaces:

  1. Get Space metadata:

    "Get details about this Space: [URL]"

  2. Via scripts/scrapeSpaces.js:

    • Find and document active/scheduled Spaces
    • Get speaker lists
    • Track topics and categories

  3. Note: Recording actual Space audio requires different tools — XActions captures metadata and participant info.

Part 10: Export Workflows

Complete export workflows for different purposes:

Archive Your Entire Account

# Profile
xactions profile myusername --format json --output archive/profile.json

# All tweets
xactions tweets myusername --limit 5000 --format json --output archive/tweets.json

# Followers snapshot
xactions followers myusername --limit 10000 --format json --output archive/followers.json

# Following snapshot
xactions following myusername --limit 5000 --format json --output archive/following.json

# Bookmarks (via MCP or browser script)
# DMs (via browser script)

Research a Topic

# Search tweets
xactions search "topic" --limit 200 --format json --output research/tweets.json

# Top voices (extract from tweets)
cat research/tweets.json | jq 'group_by(.author) | map({author: .[0].author, count: length}) | sort_by(-.count) | .[:20]'

# Save key threads
xactions thread "URL1" --format markdown --output research/thread1.md
xactions thread "URL2" --format markdown --output research/thread2.md

Content Inspiration Database

# Viral tweets in niche
xactions search "topic min_faves:500" --limit 100 --format json --output inspo/viral.json

# Bookmark exports
# Thread unrolls of best content
# Media from top creators

My Media Goals

(Replace before pasting)

  • What I want to download/export: Videos / Threads / Media / Everything
  • Specific tweets to unroll: URL1, URL2
  • Accounts to archive media from: @account1, @account2
  • Purpose: Personal archive / Research / Content creation

Start with Part 1 — show me how to download a video from a specific tweet.

⚡ Explore XActions

100% free and open-source. No API keys, no fees, no signup.

Browse All Documentation