HTTP proxy
An HTTP proxy is a proxy that speaks the web's own protocol: it reads each request's address and headers and can filter, cache or log them.
Explanation
In depth
An HTTP proxy is the classic proxy you configure in a browser or operating system with a host and port. For a plain http:// address the browser hands the whole request to the proxy, which can read the URL and headers, apply rules (block, cache, log, rewrite) and fetch the page. For an https:// address the browser instead sends a CONNECT request naming the host and port, and the proxy relays encrypted bytes it cannot read, unless the organisation has installed its own root certificate on the device so it can decrypt and inspect. Squid, corporate web filters and secure web gateways all follow this pattern. AnyProxy is not a configured HTTP proxy: there is no host and port to enter. It is a web proxy, a website that fetches pages on your behalf, but it does the same job an HTTP proxy does on the server side. One practical point matters for every kind of proxy: a site that only offers plain http:// is unencrypted somewhere along the way. Through AnyProxy, your browser to AnyProxy is HTTPS, but AnyProxy to an http:// site is not.
In practice
HTTP proxy in the real world
An HTTP proxy sees and can cache whole plain-HTTP requests; for HTTPS it only sees the CONNECT line with the host name and port. Common ports are 3128, the default for Squid, and 8080. On 15 October 2026 we opened http.badssl.com, a test page that is deliberately served without encryption, through AnyProxy's Singapore server. It loaded, but the leg from the proxy to that site carried the page unencrypted, which is exactly what its red background is there to warn about.
Sources: RFC 9110 - HTTP Semantics (CONNECT) · badssl.com plain-HTTP test page
We checked
A plain-HTTP site through a proxy
Questions about HTTP proxy
Related terms