crawl4ai

Plugin crawl4ai from marketplace crawl4ai-cloud in unclecode/crawl4ai-cloud-plugin. Commands read from the repo's manifest — 26 days ago.

Skills (7)

• bulk-crawl

  Scrape and save more than 10 URLs with Crawl4AI Cloud (from a list, a file of URLs, a sitemap or search results). Use when the user gives more than 10 URLs or a file of URLs to crawl, scrape, save or download. It shows the cost estimate, waits for the user's yes, then saves the pages under .crawl4ai

• extract-data

  Get structured, typed JSON data from a web page, such as product prices, a table, a list of articles, contact details, specs or any fields the user names. Use when the user wants data pulled out of a URL into fields, a list or a JSON shape, not the full page text.

• google-serp

  Get Google's results page as structured data with Crawl4AI Cloud, with the true rank of each result and every block Google showed (AI overview, people also ask, local pack, top stories, ads, videos). Use for rank tracking, SEO checks, competitor visibility, "where do we rank for X", "who ranks for X

• recipes

  Run a Crawl4AI Cloud recipe, a ready-made scraper that turns a known site's pages into rows (for example Hacker News "Who is hiring" posts). Use when the user asks for recipes, ready-made scrapers, or data from a site that a recipe may cover.

• scrape-to-files

  Save web pages as markdown files under .crawl4ai/ in the project, so later steps read only the parts they need. Use when the user wants to keep, save, download or archive 1 to 10 pages, wants pages as local reference or context, or when a page is long (documentation, specs, long articles). Give the

• watch

  Watch a web page for changes with Crawl4AI Cloud and read what changed, such as a price going up, a plan being removed, a new article or a product back in stock. Use when the user wants to monitor, track or be told about changes on a page, or to stop or pause a watch.

• web-research

  Answer a question about the world outside the project whose answer may have changed since your training data, with source URLs. Use it for recent releases and versions, news, current prices and plans, who runs or maintains something today, and current documentation, before you call a Crawl4AI tool,

Slash commands (7)

• /crawl4ai:balance

  Show your Crawl4AI Cloud credit balance

• /crawl4ai:crawl

  Scrape and save many URLs, or a file of URLs, with Crawl4AI Cloud

• /crawl4ai:extract

  Extract structured JSON from a web page with Crawl4AI Cloud

• /crawl4ai:feedback

  Send a bug report, a feature request or feedback to the Crawl4AI team

• /crawl4ai:scrape

  Scrape one web page to markdown with Crawl4AI Cloud

• /crawl4ai:search

  Search the web with Crawl4AI Cloud

• /crawl4ai:serp

  Google's results page as data, with true ranks, with Crawl4AI Cloud

Subagents (1)

• web-researcher

  Research a topic on the web in depth with Crawl4AI Cloud, in its own context, and return a short summary with sources. Use for research that needs many searches or many pages, such as comparing products, surveying documentation, or collecting facts from several sites, so the pages stay out of the ma

MCP servers (2)

• crawl4ai

  connects to: ${user_config.base_url}/mcp

• watch

  runs: python3 ${CLAUDE_PLUGIN_ROOT}/scripts/watch_channel.py

Similar plugins