Introducing Universal Scrape

Your agent can do a lot with the right context. Getting it that context can still mean building a scraper, parsing a file and connecting another API before you get back to the thing you wanted to build.

We've been building Firecrawl to handle that work. Today we're introducing Universal Scrape, bringing more of the data your agents need into the /scrape endpoint you already use.

Give your agents context from web pages, PDFs and images, plus people profiles, financial data, podcasts and beyond.

Access the web, files and a growing set of Alexandria providers, all with one API.

Firecrawl handles the rendering, retries and parsing for web pages and files. With Alexandria, Universal Scrape can also reach provider data through the same API.

• Web pages and files. Read dynamic websites, PDFs and images, along with Word documents, spreadsheets and presentations available by URL.
• People and company profiles. Get professional and company data through providers including Apollo, FullEnrich and Data Legion.
• Financial data. Reach company financials and market information through providers such as Fiscal.ai and Benzinga.
• Podcasts. Find relevant episodes and conversations through Particle, with tools for podcast search, lookup and transcripts.

Alexandria has more than 200 providers, and Universal Scrape already works with a growing selection of them. We're adding providers to Alexandria every day and bringing more of that library into /scrape.

If you're already using Firecrawl, you can keep scraping the pages and documents your agents rely on and add provider data when you need it. Both use your Firecrawl API key, with provider calls billed in Firecrawl credits at each tool's listed price.

For a web page or supported document, send its URL. For provider data, call a supported Alexandria capability through the same endpoint. Some supported profile URLs also route directly to your configured enrichment providers.

Here's how to read a page and find podcast conversations with the same client. The second call runs Particle's podcast search through /scrape and returns matching transcript segments.

Use the latest firecrawl SDK and set FIRECRAWL_API_KEY. Before running the podcast call, an org admin must accept Particle's provider terms.

import { Firecrawl } from "firecrawl";
 
const firecrawl = new Firecrawl({
  apiKey: process.env.FIRECRAWL_API_KEY,
});
 
const page = await firecrawl.scrape("https://www.firecrawl.dev", {
  formats: ["markdown"],
});
 
const podcasts = await firecrawl.scrape({
  alexandria: {
    provider: "particle",
    capability: "podcasts/episodes/search",
    options: {
      semantic_search: "AI agents",
      limit: 2,
    },
  },
});

The Scrape docs cover URL routing, and the Alexandria docs explain how to find and call provider tools.

Bring the data you need into whatever you're building, wherever you use AI agents. Install the Firecrawl plugin for ChatGPT and Codex, connect Firecrawl in Claude, or add the Claude Code plugin. Connect your Firecrawl account, then ask your agent to read a URL or find data in Alexandria.

You can also build with the API or connect your agents through MCP. The CLI is available for terminal workflows.

We'll keep expanding what you can reach through /scrape. The goal is for a new source to give your agent more to work with without giving you another integration to maintain.

Try Scrape · Browse Alexandria's providers · Read the docs