Supporting Redshift is actually pretty simple. The only problem is that you need to pay for a cluster for testing. However, since it's pay-by-the-hour you can keep charges minimal (don't forget to delete the cluster when you are done testing!)

Here are a few gotchas with redshift: 1) OID and CTID columns show up in the catalog, but you can't access them.

2) Primary Keys and unique indices aren't actually enforced!! You can insert two rows with the same primary key.

3) Make sure you only access catalogs that exist in 8.0, eg. there is no pg_matviews before 9.3, there's no typcategory before 8.4 etc. (fortunately the PostgreSQL docs are very helpful)

4) When setting the "application_name" connection parameter, PostgreSQL will have to connect twice to a Redshift server. Recent versions of libpq do so automatically, but you should keep this in mind (eg. when using SSH tunnels, or when validating SSL certs)

However, all in all, it's pretty easy to add support for Redshift. (I know because I'm the developer of PG Commander)