Crw

fastCRW is a web data API for AI agents that searches, scrapes, crawls, maps and extracts web content through one managed endpoint. It turns a URL into clean markdown, JSON or HTML with full JavaScript rendering, crawls whole sites into structured data, discovers every URL on a domain, pulls structured JSON with a prompt or schema, monitors pages for changes with a webhook, and converts uploaded PDFs to markdown. The vendor positions it as an open source alternative to Firecrawl and Tavily. It is aimed at developers building agents, RAG pipelines and research tools. Clients are available for Python and Node, alongside cURL and a first class MCP server for coding assistants and agent frameworks. The core engine is written in Rust and released under the AGPL version 3 license, so teams can audit it or run the single binary on their own infrastructure at no cost, while the managed cloud handles proxy rotation, browser rendering and anti bot measures. Pricing is credit based, where one credit equals one page regardless of renderer, and failed requests are not billed. Free includes 500 credits to start with up to 500 more from onboarding tasks and no card. Launch prices billed yearly are Hobby at $11 per month (5,000 credits), Standard at $59 per month (100,000 credits), Growth at $237 per month (500,000 credits) and Scale at $467 per month (1,000,000 credits). Top ups run from $9 for 1,000 credits to $299 for 500,000 and never expire. Enterprise offers custom volume, dedicated clusters and an SLA.