One endpoint in front of every scraping API. When a provider blocks, the next one runs, and the response tells you which and why.
- https://api.scraperapi.com/?api_key=KEY&url=https://example.com
+ http://localhost:8787/v1?api_key=KEY&url=https://example.comSame parameter names, same response, one hostname. Today that hostname is your own, because the only way to run Proxlane is to run it yourself.
Failover across ScraperAPI, ScrapingBee, Scrapfly and Bright Data. Providers will never fail over to their competitors; this does.
Honest success detection. Providers return HTTP 200 with a CAPTCHA in the body. Every response goes through a block detector before anything is called a success, and the response names the rule that fired.
Per-request cost, including the attempts that failed. A failover that burned two charged hops reports two.
Bring your own provider keys, or self-host. Both are free, there is no hosted endpoint, and nothing phones home.
→ proxlane/proxlane · proxlane.dev · docs
Adapters are Apache-2.0 so you can write one without inheriting copyleft, and you do not need provider API keys to contribute: the contract tests replay recorded traffic.