---
name: firecrawl
description: Use Firecrawl CLI to scrape web pages to clean markdown, search and scrape top results, crawl entire websites, or map all URLs of a domain. Use this skill whenever you need to fetch web content, scrape a URL, crawl a site, extract page text, or read documentation from the web. Prefer this over WebFetch for JavaScript-heavy pages, multi-page crawls, or when you need clean structured markdown output. Trigger on: "scrape this page", "crawl the site", "fetch content from", "read the docs at", "map all pages on", "search and scrape".
---
# Firecrawl CLI
Firecrawl turns any URL into clean markdown — handles JS-rendered pages, pagination, and full site crawls.
## Setup (first time only)
```bash
npx -y firecrawl-cli@latest init --all --browser
```
This installs the CLI and sets up browser support. Run once per environment.
## Commands
### Scrape a single page → clean markdown
```bash
npx firecrawl-cli scrape <url>
```
Use for: reading a specific page, extracting content, fetching docs.
### Search and scrape top results
```bash
npx firecrawl-cli search "<query>"
```
Use for: finding and reading multiple pages on a topic at once.
### Crawl an entire website
```bash
npx firecrawl-cli crawl <url>
```
Use for: reading all pages under a domain, extracting full site content.
### Map all URLs on a domain
```bash
npx firecrawl-cli map <url>
```
Use for: discovering what pages exist on a site before deciding what to scrape.
## Practical tips
- Start with `map` if you don't know the site structure, then `scrape` specific pages
- Use `crawl` sparingly — it follows all links and can be slow on large sites
- Pipe output to a file for large results: `npx firecrawl-cli scrape <url> > output.md`
- For JS-heavy SPAs (dashboards, docs sites), Firecrawl handles rendering automatically