diff --git a/docs/hygiene-history/ticks/2026/05/01/2105Z.md b/docs/hygiene-history/ticks/2026/05/01/2105Z.md
new file mode 100644
index 000000000..966374b92
--- /dev/null
+++ b/docs/hygiene-history/ticks/2026/05/01/2105Z.md
@@ -0,0 +1 @@
+| 2026-05-01T21:05:00Z | opus-4-7 / autonomous-loop tick | 98fc7424 | Review-fix + new-bullet tick on PR #1160 (meta-rule). **REFRESH WORLD MODEL**: 7 PRs (4 merged, 3 open). #1160 had 3 unresolved review threads — investigated per the BLOCKED-with-green-CI wake-time discipline. **Three findings + one new framing landed**: (1) Codex P2 + Copilot P0 dup — SQLSharp DI pattern bullet pointed at memory file that's on PR #1155's branch (not main); CLAUDE.md pointer would dangle when #1160 lands. Fix: remove the SQLSharp bullet entirely from #1160; defer to follow-up after #1155 merges. Each PR's pointer targets must exist on main at merge time. (2) Copilot — ".claude/rules" quoted from maintainer's verbatim message; Zeta doesn't have that directory (uses skills/agents/commands). Fix: brief inline note clarifying `.claude/rules/` IS canonical Anthropic per `code.claude.com/docs/en/memory` but Zeta's discoverable surfaces are `.claude/{skills,agents,commands}/`. Adopting `.claude/rules/` is a viable future addition. Per maintainer 2026-05-01 ("we can do whatever here to get it right IDK the right answer it's probably play around a bit") the clarification stays open-ended rather than locking in a position. (3) New maintainer framing 2026-05-01: *"it could just remind to you use the router as lookup of existing substrate, quick inventory via router"* + *"the claude.md should be sufficient for him to figure out skill router as substrate inventory"* (buddy-test design constraint: no priming). Added new CLAUDE.md ground-rule bullet — **skill router as substrate inventory before authoring**. Before writing new memory file / rule / skill / agent / doctrine, search the available-skills list AND `.claude/{skills,agents,commands,rules}/` on disk for substrate on the same topic. The router's description-keyed search IS Zeta's structured-substrate index; using it as inventory pre-authoring prevents goldfish-ontology recurrence. The bullet IS the wake-time encoding so fresh-instance buddies inherit the discipline without primed prompting. All 3 threads resolved. Pointer-target verification done before commit (only existing-on-main + landing-in-this-PR memory files referenced). Cron 98fc7424 healthy. | [PR #1160 +1 commit (34e7c78); 3 review threads resolved; new skill-router-as-substrate-inventory bullet — landed at wake-time scope so buddy-test (and normal Otto) inherit the discipline] | The buddy-test design constraint (no primed prompting; CLAUDE.md must be sufficient for fresh instance to discover the discipline) is itself a substrate-design principle: any CLAUDE.md bullet must teach future-Otto how to find substrate, not just contain the substrate. Today's worked example: instead of the meta-rule alone telling future-Otto "rules must land in CLAUDE.md or pointer," the inventory-via-router bullet teaches future-Otto HOW to discover existing substrate (via skill router + .claude/ directory listing) before authoring new substrate. Two-bullet composition: meta-rule says where lessons must land; inventory-router bullet says how to find existing lessons before writing. Maintainer's "i got an idea forming" + "play around a bit" framing — the .claude/rules adoption question stays open as exploratory. |