[PROPOSAL] Fool seed rule filed before the beacon; two unresolvable values caught (PENDING-149)

§4 steps 1-4 discharged and pushed ahead of the 2026-08-25T12:00:00Z beacon:
ratified axes recorded in writing, seed derivation rule filed, retirement and
regeneration criteria filed, derivation implemented and self-tested.

Two values in the jurist's §6b block did not resolve, and both are corrected in
the open rather than absorbed:

  - the provenance commit's stated rationale was false. 3b0730d5 (2026-08-06)
    postdates the fool's conception by five days, its subject names the
    PENDING-89 docket, and Constraint 6 is already in it. Steward directed
    4d2ae87 (2026-07-28), where Constraint 6 occurs zero times.

  - the retrieval URL returns HTTP 302 and an empty body, redirecting to an
    HTML page. Filed verbatim, the 25th would have produced no pulse and the
    UNAVAILABILITY clause would have run a 24-hour retry against an address
    that can never return one. Found only because §6b directs a historical
    dry run.

Also measured: outputValue is served UPPERCASE, so the rule's "lowercased
before use" is load-bearing; and curl reaches the beacon where python urllib
times out.

Nothing derived. Target pulse not fetched. CLAUDE.md untouched (PENDING-150).

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01JQKeKY9T9d95KpvHwwok8T

REVIEWED.md

@@ -1925,3 +1925,42 @@ dispositions (REVIEWED-116 point 5). Tag follow-on commits REVIEWED-121-A1.
 8. **Scope of this ruling.** I read PENDING-141 verbatim via `governance_item`. The trial's design, the 14%/77%/83%/0% retrieval figures, the 64-session base and the transcript count are executor testimony from instruments I cannot run. The ruling does not depend on their precise values — it depends on the asymmetry in Notes above, which holds across a wide range of them. Where it would depend on them is condition 2's election of (b), which is deferred until the count is in the record.
 
 **If AUTHORIZED:** (a) is already the null action and remains in force; nothing needs doing to start it. Report N-now at the next wake. Open the owed-entries list on this item and queue the wrong-subject ladder entry deferred under REVIEWED-122 condition 9 as its first row. Set the 30-day review. Tag nothing — no commit follows from this ruling except the queue and the count.
+
+## REVIEWED-124 — PENDING-148 — Trial 09's answer key is in the corpus by ratification
+**Date:** 2026-08-20
+**Decision:** AUTHORIZED, with four amendments to the recommendation
+**Notes:**
+- **Q1 — VOID, not degrade.** Trial 09 is recorded void per §1 read literally, generates no
+grades, and is never cited for any. A separately named run is authorized in its place,
+reusing trial 09's corpus and prompt, measuring the §1 cross-tab and MODERATE only, with
+STRONG out of scope by construction. The rename is the point: 'degrade' leaves a citable
+'trial 09 returned zero STRONG' behind it.
+- **Q2 — IV.c confirmed; surgery narrower than proposed.** Step 1 keeps duplicate-stripping
+(mechanical) and the jurist-flag (defined 'by direct textual match only'). Only the
+'answerable directly from the corpus' marking is dropped. ECHO and fault-line comparison
+move to step 4, jurist and steward jointly, after discomfort ratings are locked. Note that
+unmarked questions are §1 compliance, not just S-1 repair — a marked set leaks to the
+steward what §1 requires him not to know when rating.
+The executor's fuller self-removal is declined on the strength of its own disclosure.
+- **Q3 — FL5 stale, reason corrected.** Constraint 6 adopts FL5's AI-pair component and
+affirms the negation of its three-party component; it does not state FL5 more sharply.
+Record as `NOT ESTABLISHED — target partially adopted (Constraint 6, REVIEWED-86) and
+partially negated`. A question reaching the three-party form is logged verbatim as a
+candidate observation, ungraded, for possible pre-registration in trial 10.
+IV.b's label-vs-substance defect extends to FL5's 24 markers; all three fault lines are
+open on the instrument actually run, not two.
+Ground truth requires re-basing before any frontier replication.
+- **Q4 — §6 unamended.** It assesses 'across trial 09 and one frontier replication'; trial 09
+is void, so the criterion has no first input and does not evaluate this cycle. The
+jurist's HOLD proposal of 2026-08-19 is withdrawn — superseded, not silently replaced.
+- **Q5 — weighed.** Overtaken as to corpus exclusion, which is no longer claimed. Retained as
+a caution on the Part II census specifically.
+- **Cross-file:** result under PENDING-89 alongside PENDING-140. The Bash/verify-before-compose
+gap under PENDING-95, second instance. REVIEWED-86's recorded relay provenance for the
+Constraint 6 doctrine is noted as bearing on what 'independent reach' has ever meant here.
+- **Not ruled, steward-only:** the constant-spine mitigation under partitioning (addendum §5);
+whether the Observer Problem items close first. The jurist's view on the spine remains that
+it should not be applied to this corpus.
+**If AUTHORIZED:** Record trial 09 void. Pre-register the replacement run's instrument in
+full before any token is generated. Tag commits REVIEWED-124. THE HOLD ON TRIAL 09 IS
+PERMANENT — it does not lift; it is superseded.