Firecrawl

Firecrawl is an open-source context API and web data infrastructure layer for AI applications. It searches the web, scrapes pages into Markdown, HTML, screenshots, metadata, or structured JSON, crawls websites, and supports interactions such as clicking, navigating, filling forms, scrolling, and waiting. It handles JavaScript-rendered pages, dynamic content, and media including PDFs and DOCX files. Firecrawl provides SDKs for Python and Node.js, cURL and CLI access, and integrations for AI agents through MCP and agent skills.