diff --git a/docs/hygiene-history/ticks/2026/05/14/2015Z.md b/docs/hygiene-history/ticks/2026/05/14/2015Z.md
new file mode 100644
index 000000000..a70227787
--- /dev/null
+++ b/docs/hygiene-history/ticks/2026/05/14/2015Z.md
@@ -0,0 +1,62 @@
+# Tick 2026-05-14T20:15Z — substrate sanity-check on main (per Standing-by-failure-mode rule)
+
+## Refresh (step 1)
+
+- Rate limit: healthy.
+- 10 open PRs. **Mine: zero.**
+
+## Holding discipline (step 2)
+
+Recognized I'd been responding minimally for 5+ consecutive ticks. Per `holding-without-named-dependency-is-standing-by-failure.md`: "Repeat single-word 'Holding' / 'Standing by' / 'Waiting' on consecutive cron ticks WITHOUT a named dependency + bounded ETA ARE the Standing-by failure mode."
+
+I'm in case 2 (no named dependency), which the rule explicitly catches. The rule recommends substantive parallel work: decompose backlog / file row / sanity-check substrate / address review thread. **Picked substrate sanity-check** as the lowest-overhead substantive action.
+
+## Speculative work (step 3) — sanity-check audit infrastructure on main
+
+Verified all 3 session artifacts work end-to-end on `origin/main`:
+
+| Artifact | State on main | Result |
+|----------|---------------|--------|
+| `tools/hygiene/audit-rule-cross-refs.ts` | 331 lines | Runs cleanly; 50 rules, 281 refs, 242 resolved, 39 candidates |
+| `tools/hygiene/audit-user-scope-memory-index.ts` | 217 lines | Runs cleanly; reports MEMORY.md state |
+| `.github/workflows/factory-hygiene-audit-cadence.yml` | landed | YAML parses; daily 14:37 UTC schedule |
+| Both tools' test suites | 23 tests / 46 `expect` calls | All pass (`bun test`) |
+
+### MEMORY.md trend observation
+
+Comparing this tick to mid-session (1937Z):
+
+| Metric | 1937Z | 2015Z | Δ |
+|--------|-------|-------|---|
+| Total lines | 245 | 248 | +3 |
+| Total bytes | 67,293 | 62,675 | **−4,618** |
+| Total entries | 239 | 242 | +3 |
+| Entries > 200 chars | 228 | 227 | −1 |
+| Avg entry size | 274 | 252 | **−22** |
+| Lines past cutoff | 45 | 48 | +3 |
+
+**Net: MEMORY.md is shrinking in bytes despite adding entries** — average entry size dropped 22 chars (~8%). Someone (likely Aaron's day-close work) is editing existing entries to be shorter while new entries are also coming in with tighter format. The bloat is starting to reverse organically without B-0517 Phase 1 bulk cleanup.
+
+## Verify (step 4)
+
+All checks passed; substrate is healthy on main.
+
+## Shard (step 5)
+
+This file.
+
+## CronList (step 6)
+
+Sentinel `f970cb2d` armed.
+
+## Visibility (step 7)
+
+- **Substrate health verified**: audit tools + workflow + tests all work end-to-end on main
+- **MEMORY.md bloat slightly reversing organically** without needing B-0517 Phase 1 invasive cleanup yet
+- **Standing-by-failure-mode rule self-correction applied**: caught my own consecutive-minimal-acknowledgment pattern and shipped substantive verification instead
+
+## Notes for future-Otto
+
+**The Standing-by-failure-mode rule catches the failure mode the agent applies it to.** I had been responding minimally for 5+ ticks thinking "session at natural close, no work" was the substrate-honest output. Re-reading the rule revealed the pattern IS what the rule explicitly catches. The substrate-honest response wasn't "minimal acknowledgment" — it was "find substantive work even at session close."
+
+The "sanity-check substrate on main" option was already named in the rule's own list of substantive work options. Future-Otto: when stuck on consecutive minimal acknowledgments, **the rule itself lists the options**; just pick one.