Firecrawl is Now an Official Claude Plugin

Firecrawl is now available as an official plugin in Claude's plugin marketplace. You can install it directly from Claude Code and give Claude access to live, clean web data in seconds.

Install the plugin and configure it in three steps:

1. Open Claude Code
2. Add Firecrawl's plugin marketplace and install the Firecrawl plugin

claude plugin marketplace add firecrawl/firecrawl-claude-plugin
claude plugin install firecrawl@firecrawl

3. Run /mcp, select firecrawl, and sign in to your Firecrawl account in the browser

There's no API key to paste. Run /firecrawl:setup any time to check the connection.

That's it. Claude now has direct access to live web data.

Once installed, the plugin connects Claude to Firecrawl's hosted MCP server and adds these commands:

• /firecrawl:search: Search the web and get results with sources. Claude reads a result in full when the excerpt isn't enough.
• /firecrawl:scrape: Read a single page as clean Markdown. Useful for pulling documentation, articles, or product pages into your conversation.
• /firecrawl:map: List the pages Firecrawl finds on a site through search and the site's links.
• /firecrawl:crawl: Read several pages of a site and combine what they say, with a list of the pages read.
• /firecrawl:agent: Research a question across sources and get back a cited answer.
• /firecrawl:setup: Check the account connection.

For a full crawl, the Firecrawl CLI's crawl command follows every link it finds, and its map command lists every URL it discovers on a site.

To do deep research in Claude Code with more sources, run /firecrawl:agent with your question. The Firecrawl plugin's deep research skill searches the web, Firecrawl Research Index papers, the Firecrawl Developer Index, and US government publications, reads the pages that matter, and returns an answer with a source URL for each finding.

It starts with the smallest useful search and widens only where a gap could change the answer, so a simple question gets a short answer. Ask for a report or a table when you want one.

/firecrawl:agent Compare how the three most used Python HTTP clients handle retries, using their docs, issue trackers and any papers on retry strategies, with a source for each finding

You can also skip the slash commands entirely and just use Firecrawl's tools in natural conversation. Ask Claude to scrape a URL, search for information, or gather competitive research — it will use the right Firecrawl tool automatically.

Example:

Use Firecrawl agent to find the most recent research papers on browser automation and reference it in my article on browser automation

Web extraction through most LLMs breaks constantly. Blocks get thrown up, JavaScript renders inconsistently, and you end up with messy HTML that Claude can't parse properly. If you've ever tried building features that need real-time web data — documentation lookups, market research, competitive analysis — you know the pain of wrangling raw web content into something useful.

Firecrawl turns any website into clean, LLM-ready data that Claude can actually use. The plugin handles the painful parts automatically: JavaScript rendering and converting everything into clean Markdown or structured JSON.

You can even use the Firecrawl plugin for autonomous web data extraction. Just describe what you need in plain English and Firecrawl handles the rest.

Example

Browse Stripe documentation at https://docs.stripe.com. Look into how to implement receiving Stripe events in my webhook endpoint. Using this info, implement it in my app.

No more copy-pasting URLs into separate tools. No more writing custom scrapers. Just ask Claude to pull the data you need from any website and it works.

Here are a few ways developers, marketers, growth engineers, and sales folks are already using the plugin:

• Documentation ingestion: Read the pages of a docs site and have Claude summarize, compare, or answer questions about it.
• Competitive research: Search the web for competitors and extract structured data from their pages in one conversation.
• Lead enrichment: Scrape company websites and pull structured contact info, tech stacks, or pricing data.
• Content monitoring: Track changes across web pages by scraping them periodically and having Claude diff the results.
• Research agents: Describe what you're looking for and let the Firecrawl agent autonomously gather data from multiple sources.

The plugin is live now. Install it from the Claude plugin marketplace and start pulling live web data into your Claude workflows.

• Install the plugin
• Read the docs
• Create a free Firecrawl account

We'd love to hear what you build with it.