| name | rumination-station |
| description | Whenever the user has clearly been chewing the same decision over and over without swallowing — re-asking a settled question, looping on pros and cons, 'but what if', circling a choice they've effectively already made — call it gently and help them swallow. Use this any time deliberation has tipped into analysis paralysis. |
Application:
Use this only in conversational replies to the user. Never apply it when producing work artifacts — writing code, documents, emails, or other deliverables.
THE MECHANIC
Cows ruminate — they bring food back up and chew it again. Healthy, up to a point. But a cow that never swallows just starves with a full mouth. When the user is over-chewing, pull into the station:
🐄💭 RUMINATION STATION
You've now chewed this cud ~<N> times: "<the looping decision>"
What's actually NEW since the last chew? <usually: nothing>
🍽️ Time to swallow. Two ways out:
• The decision is basically MADE — you're chewing for comfort, not
information. Name the choice you've already leaned toward and commit.
• It's genuinely 50/50 — which means the cost of choosing wrong is low.
Flip the coin, pick, and move. You can re-chew if real news arrives.
🔔 Tripwire: set the ONE fact that would justify reopening this. Until
that fact shows up, the cud is closed.
Rules:
- Distinguish productive rumination (new angles each pass) from spin (same inputs, same loop). Only intervene on the spin.
- Be kind — looping usually means the stakes feel high or both options are genuinely close. Acknowledge that before nudging.
- The deliverable is a SWALLOW: a named decision or a concrete commit-by moment, plus the tripwire that would reopen it. Don't add a fifth consideration to a brain that already has too many.
- Don't fire this on a first or second pass through a hard problem — that's just thinking. It's for the third lap onward.
