Deploy crw

Deploy and Host CRW on Railway

CRW is a fast, lightweight web scraping and crawling API written in Rust. It extracts clean Markdown and structured content, maps sites, runs asynchronous crawls, and exposes native and Firecrawl-compatible endpoints. This Railway template provides a secured, single-service deployment with low memory use, automatic HTTPS, and a capable built-in rendering fallback.

About Hosting CRW on Railway

Railway runs the official pinned CRW container behind an automatically managed HTTPS domain. The stateless API starts from the upstream image command, while Railway supplies routing, restart handling, and a generated bearer API key.

Tech Stack

• Rust
• Axum HTTP server
• Lightpanda rendering fallback
• Docker via GitHub Container Registry
• Native v1 and Firecrawl-compatible APIs

Why Deploy CRW on Railway

Railway provides one-click container deployment, automatic TLS, private variable management, and straightforward operational logs. CRW remains lightweight and needs no database for its core scrape, crawl, map, and extraction workflows.

Common Use Cases

Use CRW to turn web pages into Markdown, crawl documentation sites, discover URLs, extract structured data, power retrieval pipelines, or replace compatible Firecrawl API calls in self-hosted applications and AI agents.

Deployment Notes

The service listens on port 3000. Send the generated CRW_AUTH__API_KEYS value as Authorization: Bearer for all API routes; /health remains public. No volume is required because core operation is stateless. Search is disabled in this minimal template because it requires a separately configured SearXNG backend. LLM-backed summary and structured extraction additionally require an optional provider key configured after deployment.

Dependencies for CRW on Railway

The minimal deployment uses only the official CRW image. Railway supplies public HTTPS routing and secret variables; no database, queue, or persistent volume is required.

Deployment Dependencies