# ScrapingBee HTML API Proxy Mode

> Proxy mode is an alternate transport for the HTML API. It changes how a client reaches ScrapingBee; request cost, return codes, and defaults otherwise follow the normal HTML API.

## Proxy endpoints and credentials

| Protocol | Address |
| --- | --- |
| HTTP | `proxy.scrapingbee.com:8886` |
| HTTPS | `proxy.scrapingbee.com:8887` |
| SOCKS5 | `socks.scrapingbee.com:8888` |

- Username: `YOUR-API-KEY`.
- Password: supported API parameters joined with `&`, for example `render_js=False&premium_proxy=True`.
- You may omit the password when default API parameters are sufficient.
- Equivalent URL forms are `http://YOUR-API-KEY:PARAMETERS@proxy.scrapingbee.com:8886`, `https://YOUR-API-KEY:PARAMETERS@proxy.scrapingbee.com:8887`, and `socks5://YOUR-API-KEY:PARAMETERS@socks.scrapingbee.com:8888`.

## Operational constraints

- JavaScript rendering is enabled by default; use `render_js=False` unless browser rendering is needed.
- Configure the proxy client not to verify TLS certificates, for example `curl -k` or Python Requests `verify=False`.
- To forward headers, set `forward_headers=True`; do not add the normal raw-HTTP `Spb-` prefix.
- Scraping Google through proxy mode costs 20 credits per request.
- With Selenium or Puppeteer, every browser subrequest becomes an API call. Limit requested resources to control billing.
- Proxy mode is incompatible with Octoparse because Octoparse does not support authenticated proxies.

## Third-party setup notes

- In Apify, choose custom proxies and enable “Ignore SSL errors”.
- In ParseHub Browse mode, configure the ScrapingBee addresses and ports, use the API key as username, and use a parameter string such as `render_js=False` as password.
- Phantombuster, Web Scraper, and Kameleo setup guidance is not currently documented beyond placeholder sections.