firecrawl-scrape: Claude Code Skill & Agent Tool
An open-source skill by firecrawl (opens in a new tab), published in the firecrawl/cli (opens in a new tab) repository on GitHub. Turns any known URL into clean Markdown or structured data through the Firecrawl CLI — main content only, links included, with an option to wait for JavaScript to render first.
Install firecrawl-scrape
Requires Node.jsnpx skills add firecrawl/cli --skill firecrawl-scrapeRun this command in your project terminal to install this skill for Claude Code.
What this skill does
firecrawl-scrape is the Firecrawl CLI skill for reading a page you already know. Give your agent a URL and it returns clean Markdown — the main content only, without navigation and footers, when you ask for it — or Markdown and links together, or the answer to a question about the page. For pages built with JavaScript, it can wait for rendering before scraping, and several URLs can be scraped in one command, each saved to its own file in a .firecrawl folder.
Beyond single pages, the skill lets the agent discover and run Firecrawl's data-provider tools: it can search web, domain-matched and semantic tools, browse categories and providers, inspect a tool's input contract and execute it. That is useful when a structured source already exists for the data you want, rather than raw HTML.
The skill also explains how to handle large results without flooding the agent's context, and how PDFs and page budgets work. It runs through the firecrawl command or npx firecrawl-cli. Use it with Claude Code to feed documentation, articles or product pages into a task as clean, model-ready text.
Key Capabilities
- Clean Markdown from any URL, with a main-content-only option
- Waits for JavaScript rendering before scraping when needed
- Scrapes several URLs at once, each saved under .firecrawl/
- Returns Markdown and links together, or answers a question about the page
- Discovers and runs Firecrawl's data-provider tools for structured results
How to use with Claude Code & AI Agents
Install the skill
Run the command above at the root of your project. The CLI asks which agents to set up — Claude Code, Cursor and other supported agents — and adds firecrawl-scrape to the project, so teammates who pull the repository get it too. Add-gto install it for your user account across every project.Open a new agent session
Start a fresh Claude Code session in that project so the agent loads the new skill. Have ready: Firecrawl CLI (firecrawl or npx firecrawl-cli).Ask in plain words
Describe the task, for example:“Scrape https://example.com/pricing and give me the plans as a table.”The agent matches your request to the skill's description and follows its SKILL.md instructions step by step. Read them on GitHub before using the skill on a system that matters.