Field Notes: Running a Research Vault

Treasure Among Us treats a treasure hunt the way an analyst treats a messy dataset: the raw material is interesting, but it is worthless until it is organized, graded, and navigable. These notes describe the working habits the vault is built to support.

Why a vault, not a notebook

A notebook is linear. A real hunt is not. You are holding a poem, three competing map theories, a pile of historical sources, and a rabbit hole you are not ready to abandon — all at once. The vault exists so none of those threads have to be deleted to make room for the others.

Keep the thread. A theory you abandon today is a source you need next month.

Grading sources

Every fact in the vault carries a grade so a confident leap is never confused with a verified anchor:

• Anchor — independently verifiable; safe to build on.
• Working — plausible and useful, not yet confirmed.
• Rabbit hole — kept for completeness, flagged so it does not quietly

become load-bearing.

The daily loop

The vault is designed around one repeatable loop:

• Capture new material into the right hunt and box.
• Grade it before it touches an existing theory.
• Link it to the poems, hints, places, and maps it actually affects.
• Leave a journal entry so future-you knows *why* the link was made.

Built to scale past one hunt

The structure is multi-hunt on purpose. TTI is the first hunt the vault holds, but markdown includes, box configuration, and gated previews mean a second or third hunt drops in without a rebuild — the research method travels even when the puzzle does not.