Trillion

Say whether it is known, new or re-checked

Before answering any question about what exists, what was decided or what was found, read everything already held about it, then label every claim as known before (since when, linked to where), a new finding, or re-checked now (still true or changed). Write new findings back where the next reader looks first, and give the link to any screen that already shows the answer, so nobody rediscovers what was already known and the reader can tell a first discovery from a re-check.

What it asks for

Nothing: your AI starts it with your words.

The steps your AI walks

  1. Read what is already held
  2. Label every claim
  3. Write new findings where the next reader looks first
  4. Answer with the labels and the screen

Use it from your AI

Connect the AI you already use to Trillion once. Then ask in your own words; your AI finds this workflow by what it does and runs it.

Say to your AI: “Use Trillion to before answering any question about what exists, what was decided or what was found, read everything already held about it, then label every claim as known before (since when, linked to where), a new finding, or re-checked now (still true or changed). Write new findings back where the next reader looks first, and give the link to any screen that already shows the answer, so nobody rediscovers what was already known and the reader can tell a first discovery from a re-check”

What your AI calls: start_workflow {"workflow_id":"workflow:say-if-it-is-known-new-or-rechecked"}