API that scrapes live sites into clean Markdown for AI agents
AGPL-3.0
- TypeScript
- Python
- Rust

About Firecrawl
Firecrawl is a crawler and scraper that turns live web pages into clean Markdown or structured data for AI agents and LLM apps. It finds sources, extracts content, and handles the hard parts of crawling so you do not have to build your own browser stack.
Scrape, crawl, search, and map endpoints cover single pages through whole sites, including JS-heavy pages and media parsing for PDFs and DOCX. Actions can click, scroll, write, wait, and press before extraction, and output comes back as Markdown, structured JSON, or screenshots.
Firecrawl is open source under AGPL-3.0, with SDKs for Python, Node.js, Java, Elixir, and Rust. It can be self-hosted or used as a hosted service, with extra features in the cloud version.
Key features
- Scrape pages into clean Markdown or JSON
- Crawl whole sites and map their URLs
- Search the web and return page content
- Actions to click, scroll, and type before extract
- Parse JS-heavy pages, PDFs, and DOCX
Details
- First released
- 2024
- Deployment
- Self-hosted · Hosted service
- Output
- Markdown · JSON · Screenshots
- SDKs
- Python · Node.js · Java · Rust
- Language
- TypeScript
- License
- AGPL-3.0
