Are the header names standard?

Yes. Every header name comes from a built-in table of standard, RFC-defined HTTP headers such as Content-Type, Cache-Control, and Authorization, split into request and response sets.

The values are plausible, format-correct samples generated to match each header, such as a Bearer token shape for Authorization or a valid date for Date. They are fabricated for testing, not captured from real traffic.

What is the difference between request and response headers?

Request headers are sent by the client and describe the request or the client, such as Accept and User-Agent. Response headers are sent by the server and describe the response or the server, such as Content-Type and Set-Cookie. Some headers can appear in both.

Can I paste the output straight into a tool?

Yes. The output uses the standard Name: value format, one header per line, so you can paste it into curl config, a REST client, or a mock server definition with minimal editing.

Why generate random headers?

Randomized but valid headers are useful for fuzzing parsers, populating mock responses, and demonstrating how middleware reacts to a range of headers, instead of testing against only the handful you wrote by hand.

Random HTTP Header Generator

Generate realistic HTTP header lines — request, response, or both — using standard, RFC-defined header names paired with plausible, format-correct values. Handy for mock servers, parser fuzzing, and documentation.

How it works

An HTTP header is a Name: value pair on its own line. This tool keeps separate tables of standard request and response headers, then for each chosen header it generates a value that matches that header’s expected shape:

Authorization: Bearer <random token>
Content-Type:  <random MIME type>
Date:          <current GMT date>
Cache-Control: <random directive>

Names are real and standard; values are fabricated but format-correct, so the output parses like genuine traffic without exposing any real data.

Tips and notes

Header names are case-insensitive per the spec, though the canonical capitalization (such as Content-Type) is conventional and used here. When feeding the output into a strict parser, keep the single space after the colon and one header per line — both are what well-behaved clients and servers emit.