SCOUT Provision: Point the Machine at a New Product

Before any marketing work can be checked, something has to establish what's true about the product. This reads the product where its proof lives, its repo, its data, its artifacts, and writes three files: truth, voice, and an evidence inventory.

Section 3 is the interesting one, because it writes three files and refuses the fourth. The refusal is deliberate: the missing file is the one that can't be written honestly from the product alone. Section 4 covers taking the inventory honestly, including recording what evidence you don't have, which is the half most inventories quietly skip.

Floors apply, and the closing section names what it doesn't decide.

It ships evidence_lint.py and a provision checklist.

This is one of the six SCOUT skills. It reads and writes the shared .monkeys/ pack, so it works on its own and works better alongside the rest.

Where this came from

Deliberon is a Windows desktop app that runs a council of AI agents on a hard decision and hands back a Decision Record: dissent preserved, every claim marked as proved, asserted, or estimated.

Provision is step zero: it is what I run before the machine is allowed to say anything about a product.

Pay once, it's yours, no subscription. Thirty days of the full council with no account and no card. Runs on your machine with your own model keys.

https://deliberon.com

The rest of my work, organised by material: https://ryan-heltemes.com