REFUSED

AI Comment Prompt Leak Detector

56/130 FLOOR 65

Recorded 2026-06-14 · refused 2026-08-01

THE IDEA, IN FULL

Tool to detect and auto-fix AI-generated code comments that leak the original prompt or contain unnecessary temporal references. HN post about vibed code leaving behind comments that reveal what was asked, making code reviews harder.

WHY IT WAS REFUSED

Refused at 56 out of 130. The machine's floor is 65, so this one never reached a shelf. Where it lost, in the gauntlet's own words: Willingness to pay scored 1 of 15: "No explicit willingness to pay."; Pain intensity scored 7 of 20: "Low urgency, likely a 'nice to have'.". HN Algolia items API returned: "Is there a name for the type of comments agents add where they leak the prompt?", 18 points, posted 2026-06-14.

THE SCORECARD

Pain intensity 7/20
Market depth 10/15
Willingness to pay 1/15
Competition 9/20
Timing 8/20
Business shape 11/20
Buildability 10/20

THE SOURCE

https://news.ycombinator.com/item?id=48522763 reachable checked 2026-07-31

One source, opened by our own checker on the date shown. If it dies, this line stops saying reachable.

Free to take. Build it. We will not sell it.

There is no catch and no form. This idea is off our shelf permanently, so there is nothing for us to protect. If you build it, you owe us nothing.