We just shipped Webhooks for /agent 🚀 Subscribe to agent events and get notified after every scrape, search, completion, or failure. You can even track credit usage as it runs. Check out the docs 👇
About us
The Web Data API for AI
- Website
-
https://www.firecrawl.dev/
External link for Firecrawl
- Industry
- Technology, Information and Internet
- Company size
- 11-50 employees
- Type
- Privately Held
Employees at Firecrawl
Updates
-
Introducing Hack the Stackathon 🔥 A one-day hackathon cohosted by Firecrawl, Resend, and Reducto at YC in SF. Thousands of dollars in perks. Limited spots. Apply by Jan 28th 👇 https://lnkd.in/edwH6trR
-
-
All the formats in our /scrape endpoint explained 🧵 The formats parameter lets you mix and match outputs to fit your workflow. Get markdown, HTML, screenshots, links, and more from a single call. 1/ Markdown - The gold standard for LLMs. We return clean, structured markdown that gives your AI agents the full context of a page without the noise. Perfect for RAG pipelines. 2/ Summary - Skip the extra LLM step. Instead of scraping and summarizing manually, Firecrawl digests the page content for you. Get an instant, AI-generated summary of any URL. 3/ Links - Need to map out where a site pushes people? This format returns every single link found on the page. Simple, fast, and essential for building custom crawlers. 4/ HTML - The source of truth. Get the actual HTML of the page. You can choose between raw for the original code or cleaned to strip away the clutter. 5/ Screenshot - Capture the visual context. - Viewport: Exactly what the user sees on load - Full Page: The entire page from top to bottom Perfect for visual verification. 6/ JSON - Turn websites into structured APIs. Define a specific schema and a query, and we’ll extract exactly what you need. No more parsing messy HTML manually. 7/ Branding - Extract a website's DNA in one shot. Get logos, color schemes, fonts, and design styles instantly. Ideal for coding agents that need to clone or match existing aesthetics. Try out any of our 7 formats today! 🔥
-
Firecrawl reposted this
Agent can now take screenshots 🎇 Ask @firecrawl agent to take screenshots of whatever you need on the internet Stay tuned. A LOT more updates coming 🫡
-
-
Introducing our n8n integration v2.0 🔥 Here's what's new in this version: 1/ Added the Agent node with sync and async modes. Describe what data you want and it searches, navigates, and gathers it from even complex websites. Works for single datapoints or entire datasets at scale. https://lnkd.in/ejMxvYWW 2/ Nodes are now grouped by resource type. Agent Actions, Map & Search, Crawl, Account, and Extract (Legacy). Note: Existing workflows will need the resource type manually selected after updating. 3/ Schema generation is more flexible now. You can paste a JSON example of what you want and it builds the schema for you, or write the schema manually if you need full control over the structure. 4/ Batch scrape now takes URLs in a single text area instead of individual fields. This means the node can properly receive multiple URLs from previous nodes in your workflow, added manually or passed dynamically. 5/ All Firecrawl nodes now work as tools inside n8n's AI Agent with full $fromAI() support. Your agent can autonomously scrape, crawl, and search while defining parameters like URLs and prompts dynamically. That's it for major changes, just update your n8n instance to try it out!