FirecrawlUnclaimed

Firecrawl is a web data API. It searches the web, scrapes pages, crawls sites, and interacts with pages, then returns the content as clean markdown or structured JSON. The page describes it as a way to give AI agents web data, and it is also available as an MCP server and a CLI.

The API covers search, scrape, map, crawl, and interact. Scrape returns a page as markdown, JSON, or a screenshot. Interact lets an agent click, type, scroll, and navigate a page using AI prompts or code. The page says Firecrawl handles JavaScript rendering, waiting for content to load, and parsing PDFs and Word documents.

The product is aimed at developers and teams building AI agents, chat assistants, research tools, and onboarding or lead enrichment features. The site lists use cases such as deep research, AI chat with live web content, and sales data enrichment.

The code is open source on GitHub. Developers can call the API from Python, Node.js, cURL, or the command line, or connect an AI agent through MCP.

Pricing has a free plan with 1,000 credits a month and no card required. Paid monthly plans start at Hobby for 5,000 credits, at $19 a month billed monthly or $16 a month billed yearly. Standard, Growth, and Scale plans add more credits, concurrent requests, and support. Enterprise pricing is custom. Each credit covers one basic page scrape or crawl, and search costs 2 credits per 10 results. Failed requests with no result are not charged.