Agency under uncertainty

Most tools assume you already know what to do and just need to organize it. Usually that is false. The Assumption Registry turns each consequential belief into a record with an honest status, the evidence for and against it, the cheapest test that could change it, and the decision it endangers — so your most dangerous unknown becomes visible, testable, and killable.

AI may organize evidence. It may not manufacture evidence.

What is inside

• Assumptions — every belief with its honest epistemic status, confidence, consequence-if-false, and a Stale flag past its review date.
• Evidence — what was observed, its strength and direction; interpretations stay on the assumption.
• Experiments — a Valid flag refuses any test without a pre-written disconfirming result.
• Decisions — the choice being protected, standing on its assumptions; killing one keeps its record.

<aside> 🗂️ Universal mode: works with no paid Notion AI. Sort Assumptions by Consequence if false + Confidence to surface your most dangerous unknowns. Example records are tagged — replace them with your own.

</aside>

The working instrument

Assumptions

Evidence

Experiments

Decisions

Using it

The loop is small and complete: a desire becomes an assumption, an assumption meets evidence, the weakest belief becomes an experiment, the result becomes judgment, and judgment becomes a commitment you can defend — or a fantasy you abandon in time to save a year.

Your first ten minutes