session 2026-07-18: REVIEWED-58+63 placed — first 6 Loeb graduated (Block 1 closed; PENDING-63 full lifecycle in-day); S3 session file + ledger + harvest register + 3 KG lines

claude/memory/skill-harvest-register.md

@@ -8,7 +8,7 @@ metadata:
   node_type: memory
   type: reference
   originSessionId: a8eb1d46-314e-4545-a133-f747c69ad5b4
-  modified: 2026-07-18T13:50:27.838Z
+  modified: 2026-07-18T17:00:35.719Z
 permalink: claude-memory/skill-harvest-register
 ---
 
@@ -522,6 +522,14 @@ The single place proposed skills live so they don't evaporate between sessions.
 
 *(One candidate, jurist-minted and immediately reusable — Sessions 2 and 3 are structurally this instrument. Also this wrap, applied directly as memory-keeping, not skills: [[feedback-verify-before-compose-hook]] gained the pre-hook-artifact note [first gated edit must itself carry the GROUNDED-IN payload]; chamber `CLAUDE.md` brought current mechanically [2b status line + the `app[]`-vs-channel term]. No skill create/patch/retire beyond the one proposal.)*
 
+### Harvest 2026-07-18 (S3/evening — first graduations; PENDING-63 full lifecycle in one day)
+
+| Element | Kind | One-line | Where it lands | Status |
+|---|---|---|---|---|
+| **sandbox-must-pin-the-shared-module** | verification-ladder entry | When a test sandboxes module-level state (paths/constants), the patch must land on the SAME module object the code-under-test imports (sys.modules) — NOT a freshly-exec'd copy (a `_load()`-style loader builds a different object, and the mispinned sandbox silently runs the driven code against REAL data). Pair: any test that can touch a shared real store gets a **before/after census of the real substrate** as the standing guard. Earned hard 2026-07-18: the PENDING-63 ratification test's own first red — the gate-driven leg ran against the real Chamber Sources from inside the fleet; only the refusal path kept it read-only (censused untouched before+after). The witnessed red DETECTED — this is also the S2 watch-list item earning its instance. | `reference-verification-ladder.md` | **PROPOSED** |
+
+*Reinforcements recorded, not re-proposed:* **implementation-is-a-second-gate** (S1 proposal) is now thrice-earned and effectively jurist-endorsed — the PENDING-63 ruling's own opener: "Every significant finding here came from *doing* rather than reading… exercise-before-ratification making its own case." **witnessed-red** (S2 watch-list) earned its live instance and folds into that ladder entry as planned. **`/jurist-package`** (07-17 proposal) proved twice more today — both rulings engaged every question the packages posed; the repo-blind-brief method is now 4-for-4.
+
 ### Harvest 2026-07-18 (S2 — the 2b wiring build)
 
 No new skill proposal. Two register notes: **(1)** the S1 proposal **implementation-is-a-second-gate** (ladder entry, awaiting steward) gained its **first code-level instance** this session — the line-ref coordinate contract sat invisible through every paper gate and surfaced only when `write_sidecar` had to decide what a hash binds; the proposal is now twice-earned in two sessions. **(2)** Watch-list, not yet a proposal: **witnessed-red discipline** — "test-first" without an observed red state is fixtures-first; either run the fixtures once against the unbuilt seam, or name the evidence honestly as known-bad-fixture teeth (self-caught at S2 wrap; KG drift-pattern logged). If authorized, likely folds into the implementation-is-a-second-gate ladder entry rather than standing alone.