Anycrawl Social Extractor

AnyCrawl Social Extractor

Use this skill when the user gives a social URL and does not want to choose between the TikTok, Instagram, or YouTube skills manually.

Workflow

1. Accept a single public URL.
2. Run the router:

node .claude/skills/anycrawl-social-extractor/scripts/run.js "<social-url>"

3. Read the compact JSON summary from stdout.
4. Open the saved artifact file if the user needs the full extraction.
5. Summarize the meaningful fields for the user instead of dumping raw HTML into the chat.

Routing Rules

• TikTok URLs route to anycrawl-tiktok-scraper
• Instagram URLs route to anycrawl-instagram-scraper
• YouTube watch or youtu.be URLs route to anycrawl-youtube-video-extractor

If the URL is not one of those platforms, fail fast and tell the user it is unsupported.

Output Handling

The router does not scrape directly. It delegates to the platform-specific canonical extractor and returns:

• platform
• routedSkill
• runnerScript
• outputPath
• the delegated script's summary fields

Read references/router-notes.md if you need the exact platform mappings.