Can it gather data from several sources at once?

Yes. It includes an AI agent for autonomous multi-source data gathering; you describe what you need.

Does it work on whole sites or just single pages?

Both. It can scrape single pages and also crawl entire sites.

firecrawl

PackWeb & browsing

Firecrawl is a plugin for web scraping and crawling. It turns websites into clean, LLM-ready markdown or structured data, so an AI agent can read and work with web content. It can scrape single pages, crawl entire sites, search the web, and extract structured information, and it includes an AI agent for autonomous multi-source data gathering.

Unavailable. Delivery for this kind is on the roadmap — not serving yet.

Have access to Firecrawl, since the plugin is powered by the Firecrawl service.

What your AI can do with it

Scrape single web pages into clean, LLM-ready markdown
Crawl entire sites and convert them to markdown or structured data
Search the web from within an agent session
Extract structured information from web pages
Run an AI agent that gathers data from multiple sources autonomously

Getting started

Have access to Firecrawl, since the plugin is powered by the Firecrawl service.
Add the Firecrawl plugin to your agent setup.
Describe what you need, such as a page to scrape, a site to crawl, or data to extract, and let the agent carry it out.

Signals

GitHub stars: 37k
Forks: 4k
Last commit: Sep 2026
Hacker News mentions: 9

Questions

What formats does it output?: It turns websites into clean, LLM-ready markdown or structured data.
Can it gather data from several sources at once?: Yes. It includes an AI agent for autonomous multi-source data gathering; you describe what you need.
Does it work on whole sites or just single pages?: Both. It can scrape single pages and also crawl entire sites.

Advanced

Item type: plugin
Key: anthropics-claude-plugins-official-firecrawl
Source: github.com/anthropics/claude-plugins-official

github.com/anthropics/claude-plugins-official