Version. 0.1
Date. 2026-08-20
Rules. Ruling 1, Operator, 2026-08-20 — on the six-entry trace at loomworks-foray-r2-forward-gap-scoping-note-v0_3.md §7.3-class findings and the prior Step 0 census.
Scope. Disposition record only. _ANCHOR_PRIORITY in src/loomworks/foray/anchor_priority.py is not touched. The rename, retirement annotation, and any other registry-code change land with the R-2 CR, not here.
Governing seam. These are Loomworks-side registry-content decisions. None require FORAY's word; FORAY's own response (foray-response-to-loomworks-census-v0_1.md §2) already said as much — "whether every live kind needs a registry entry is a Loomworks design question about what that registry is for."
agent_registered — RETIRE, superseded
Ruling. Retired, not deleted. agent_identity_minted (W1, pre-registered under the current identity model) is the same event under its current name — agent_registered is the Phase 62 name for it, predating the design that settled the current agent-identity model.
What the R-2 CR does with this ruling. The registry entry stays in the dict as a retired/superseded marker (not removed outright — the trace record is the reason it isn't silently deleted), annotated to point at agent_identity_minted as its successor. The CR is where the actual annotation text and code shape land; this note records only that the disposition is settled.
induction_cycle_recorded — RENAME the entry to induction_cycle
Ruling. The registry entry is renamed to match the live writer. The recorded fact does not move — a live writer already exists and already writes event_kind="induction_cycle" at two call sites (engagement/creation.py:709, persons/personal_engagement.py:443); the registry was misnamed relative to what was already being written into memory_events rows. The writer is not renamed. The registry moves to match reality, not the other way around.
What the R-2 CR does with this ruling. Renames the dict key from induction_cycle_recorded to induction_cycle, keeping its "standard" priority value, with a comment noting the correction and citing this record.
candidate_engagement_discarded — HOLD, undisposedRuling. Not settled. Its disposition follows Ruling 2 item A below (the candidate-discard-deletes-Memory-rows design question) and must not be decided ahead of it. No action in the R-2 CR for this entry until item A resolves.
contribution_credential_issued / _claimed / _revoked — KEEP all three, annotatedRuling. All three entries are kept as-is. Removing them would erase the only surviving evidence that anyone judged these three events significant enough to register — the acts themselves are real (an authority act each: issuance, claim, revocation of a contribution credential) even though nothing currently writes a Memory event for any of them.
Annotation required (lands with the R-2 CR): each entry gets a comment stating that the contribution-credential lifecycle persists via raw SQL directly into its own contribution_credentials table (authority/contribution_credentials.py) and never calls append_event — these three acts touch neither the Memory event log nor any FORAY reserved location. This is the same fact Ruling 2 item B below files as its own open item; the annotation here is the registry-local pointer to it, not a duplicate resolution.
| Entry | Disposition | Depends on |
|---|---|---|
| agent_registered | RETIRE, superseded by agent_identity_minted | — |
| induction_cycle_recorded | RENAME → induction_cycle | — |
| candidate_engagement_discarded | HOLD | Ruling 2, item A |
| contribution_credential_issued | KEEP, annotated | Ruling 2, item B |
| contribution_credential_claimed | KEEP, annotated | Ruling 2, item B |
| contribution_credential_revoked | KEEP, annotated | Ruling 2, item B |
DUNIN7 — Done In Seven LLC — Miami, Florida Loomworks — anchor-priority registry: six-entry dispositions — v0_1 — 2026-08-20