WWID

Last checked 3 Nov 2026Template updated 3 Nov 2026

WWID (What Would I Do) is a second brain that learns how you make decisions and write, from your own messages to your other bots. It scores its predictions on your real decisions every week so you can see whether it's converging.

Memories2Facts it already knows

Lock every calibration prediction in memory before the owner answers, then log hit, partial or miss, what was wrong, and the running tally.

Common modeling error to watch for: over-predicting urgency and process. Check logistics (where they'll be, whether they're at their desk) before inferring a strategic motive.

Skills1Playbooks it can run

wwid-getting-started

Use on the first conversation with a new owner of a WWID (What Would I Do) second-brain bot.

Routines3Jobs that run on their own

Nightly WWID learning pass

Fires nightly to learn from the owner's own messages to their other bots that day.

Daily WWID calibration call

Fires each weekday to test one prediction against a real decision of the owner's.

Weekly WWID hit rate

Fires weekly to report how accurately WWID is predicting the owner's decisions.

Search this directory from your AI assistant

Add one MCP server and ask Claude, Cursor, or Grok for the right bot. Public endpoint, no API key.

Claude
Cursor
Grok

Setup guide

WWID

wwid-getting-started

Nightly WWID learning pass

Daily WWID calibration call

Weekly WWID hit rate

You may also like

Urantia Bot

Artsy Abby

All Art Calls

Travel Agent

Grove Home