Firecrawl

Firecrawl is a web data API that turns live pages into clean Markdown and structured data for AI agents and applications. Four core operations cover most work: search the web and return the full content behind each result, scrape a single page into Markdown, JSON or a screenshot, map a site to discover its URLs, and crawl it in bulk. A newer interact operation scrapes a page and then drives it with prompts or code, clicking and navigating as needed. It is built for developers wiring the live web into language model pipelines rather than for analysts working in a dashboard. Official clients cover Python and Node, with a CLI and plain cURL, and the service connects to agent frameworks and MCP clients. The project is open source. A monitor feature runs repeat checks on pages using a deterministic extraction engine, and output formats include JSON, question answering and highlights for teams that want structure rather than raw text. Pricing is credit based: one credit equals one page on a basic scrape, crawl or map, search costs 2 credits per 10 results, and interact costs 2 credits per browser minute. Free gives 1,000 credits a month and 2 concurrent requests with no card. Hobby is $19 per month billed monthly or $16 per month billed annually, for 5,000 credits and 5 concurrent requests. Standard is $83 per month billed annually for 100,000 credits and 25 concurrent requests, Growth is $333 for 500,000 credits and 50, and Scale is $599 for 1,000,000 credits and 100. Enterprise is custom and adds dedicated support with an SLA, zero data retention and single sign on. Paid plans can enable usage billing that adds credits in $5 increments up to a monthly cap you set.