I've run into few issues:

1. Performance: lot of reflection and allocations, so not the fastest thing ever.

2. Case insensitivity: https://play.golang.org/p/LtwChO_tp0 this can be pretty unpleasant when it bites you

3. Unicode: It replaces invalid utf8 with unicode questions marks (dunno how exactly is the replacement character called) but does not provide any API to detect that it happened. So that can lead to silent corruption of the data.

I think it’s well demonstrated to be slow (ns/op) and wasteful (allocs/op) in benchmarks, especially in bigger payloads.

The HTTP library is similar.

Most 3rd party libs make some trade offs that are pretty reasonable to achieve superior perf, but it’s worth understanding them to make sure it’s best for your use case.