Every corpus source carries an epistemic tier — peer-reviewed, preprint, practitioner, or ai-generated — recorded at ingest time and surfaced by consult, so that the trust level of a source travels with the source rather than living in the reader's head. The convention serves evidence discipline: AI-generated sources support nothing on their own and may only corroborate claims grounded elsewhere, and a consult result that hides tier invites laundering weak evidence into strong claims. Detector: the ledger lint refuses influence edges whose source lacks a tier; acceptance is an ingest path that cannot register a source without one and a consult output that displays it.

Every corpus source carries an epistemic tier — peer-reviewed, preprint, practitioner, or ai-generated — recorded at ingest time and surfaced by consult, so that the trust level of a source travels with the source rather than living in the reader's head. The convention serves evidence discipline: AI-generated sources support nothing on their own and may only corroborate claims grounded elsewhere, and a consult result that hides tier invites laundering weak evidence into strong claims. Detector: the ledger lint refuses influence edges whose source lacks a tier; acceptance is an ingest path that cannot register a source without one and a consult output that displays it.

Remedy grounds (2026-09-29)

• Why: the owed question's premise that ingest lives outside the binary is stale; source add and source ledger are verbs (internal/core/source/add.go), so the tier has a home at registration.
• Sources (consulted 2026-09-29): the GRADE handbook rates the certainty of a body of evidence per outcome, not a single source (https://gdt.gradepro.org/app/handbook/handbook.html), so a per-source field records publication status rather than a GRADE level.
• Rejected: GRADE's four certainty levels, and checklist scores that are a reader's judgement per use rather than a property of the source.