Pitfall library

Where these pitfalls come from, and where they go

1. 1An install fails → you hit “Install failed?” on the entry page (no login needed)
2. 2Deduped and triaged → missing dependency / blocked by permission / network blocked / version mismatch
3. 3A curator verifies it and writes it back to that entry (unverified text never reaches a prompt)
4. 4Next time anyone copies the prompt, this pitfall rides along — their AI avoids it

We are also a "try it → report back → reach the author" path

Many entries come from individual developers with no name recognition. Our job is not to hand you a verdict — it is to make trying them cheap, and to route what you find back to the author.

1. ① You find it here — every entry carries its official source and author. No proxying, no re-hosting.
2. ② You install and try it — copy the prompt into your own AI; you see where it lands and whether it worked.
3. ③ Anything you hit splits two ways — pitfalls that can be written down go here (the flow above); bugs in that software go to its official tracker, which is where the author actually reads them.
4. ④ We relay, we do not arbitrate — no judging authors, no ranking, no answering on their behalf.

Rationale: to keep unknown developers’ work tryable, we must not rank — the moment we rank, the ones nobody tried stay untried forever.