If my agent can’t use it,
I probably won’t either.

We had AI agents try to actually use well-known AI products. No humans helped. Here’s what happened.

Agent-friendly01

Recovered from a bad request on its own. Clean.

No door for an agent to even knock on.

No public API — agents are locked out by design.

Charged the same order three times. Never noticed.

Every verdict is backed by a real transcript — tap to see it.