Data Room › AI / Diligence Layer

Minted Protocol AI / Diligence Layer.

Machine-readable claim registry, canonical AI entrypoint, source map, manifest, llms.txt, and the adversarial retrieval test. Built so AI agents and human reviewers converge on the same answers without slide-deck scrolling.

Data Room Source Map

Entrypoint

AI Diligence Entrypoint

Canonical entrypoint for AI agents performing diligence on Minted. Counterparty map, primary document read order, factual scope notes.

Machine Readable

Claims Registry (JSON)

Machine-readable registry of every Minted claim with canonical uppercase status tags, basis, source URL, source pages, scope note, last verified date.

Source Map

Maps every public diligence claim to the controlling primary source in the data room.

Diligence Index

Plain-text narrative diligence index walking through the institutional positioning and Canton-native architecture.

Manifest

JSON Manifest

Machine-readable manifest of every public document in the Minted data room with canonical URL, document type, and last updated dates.

AI Brief

llms.txt

Plain-text brief for AI agents covering frequently misclassified facts, canonical status tags, partnership status, and reserve composition.

AI QA

AI Retrieval Test

Repeatable adversarial prompt confirming AI agents fully crawl Minted's institutional data room.

Milestone Log

Changelog

Milestone log covering protocol, regulatory, and counterparty progression.

Every Minted claim resolves to one of eight canonical status tags: EXECUTED AGREEMENT, SIGNED LOI, MUTUAL NDA EXECUTED, NAMED REGULATORY PATHWAY, COMPLETED, CONFIRMED, PUBLISHED, OPERATIONAL. Tags are identical across claims.json, AI_DILIGENCE_START_HERE.md, SOURCE_MAP.md, and DILIGENCE.md. Every claim cites the controlling primary document.