
by Firecrawl
Firecrawl is a web data API that searches, scrapes and interacts with web pages and returns clean, LLM-ready data. It is open source and works with AI agents through MCP and a CLI.
APIUsage-based
Step 1: Sign up for a free account and get an API key.
Step 2: Install an SDK, such as firecrawl-py for Python, or use cURL.
Step 3: Call scrape on a URL to get markdown or JSON.
Step 4: Use search, map or crawl for broader jobs, or interact to operate a page.
Step 5: To use it from an agent, run the init command for firecrawl-cli to connect through MCP or the CLI.
Run npx -y firecrawl-cli@latest init --all --browser. This connects your agent through MCP or the CLI and skills, and the site lists Cursor, Claude Code and Windsurf as examples.
Credits are consumed per API request and vary by endpoint and feature. A basic scrape or crawl uses 1 credit per page, and a scrape that returns no result is not charged.
On paid plans you can enable pay-as-you-go, which adds credits up to a monthly limit you set.
Yes. It handles JavaScript rendering and waits for content to load, and it can parse PDFs and DOCX files.
ExaExa is a search API for AI agents that returns web data, page contents and structured research results. It offers search, contents, agent and monitor endpoints.
Perplexity Search APIThe Perplexity Search API returns real-time, ranked web results as JSON, with page text extracted into each result. It is for developers who want raw search results to feed their own models, agents or pipelines.