SnipTray logo
Research

What am I missing?

Argue against your own conclusion.

  • critique
  • blind spots
  • decisions

The prompt

I have concluded: {{conclusion}}

My reasoning: {{reasoning}}

Argue the other side properly. Name the assumptions I have not examined, the
evidence I would need to be wrong, and the failure mode I am least likely to
notice.

Do not hedge into "both have merit". If the conclusion survives, say which single
piece of reasoning is carrying it — that is the part worth double-checking.

What to fill in

{{conclusion}}
What you've concluded — e.g. we should sync prompts through the existing CloudKit zone
{{reasoning}}
Your reasoning — e.g. one zone to fetch, tombstones already handled

What it produces

Unexamined: that one zone means one fetch. It also means one
failure domain — a prompt-record schema error stalls clip sync
too. You haven't priced that.

To be wrong you'd need: prompt volume high enough that zone
fetches get slow, or a prompt-side error that blocks clips.

Least likely to notice: tombstone collision between the two
record types in the same zone.

Carrying the conclusion: "one zone to fetch". Check that.

117 prompts, in the app you already paste from

The whole library, searchable, with your own prompts alongside it. Free to browse here; Pro to keep them a keystroke away on every device.