[CLAUDE] Docs: S125-S126 closeout GOP — H23-precedence DO (spawn-param THANG hard-pin, /fable-real SONG) + sweep all-inherit va 15+3+5 + gotcha #83 + hub-mail h23-result

- NHOM A restart-verify: A1 tier:'fable' tren reviewer pin-opus -> resolved claude-fable-5 (spawn-param THANG); A2 no-tier -> claude-opus-4-8 DU lead=Fable (pin an runtime); A3 autoMemory WRITE-path OK (project_h23_spawn_precedence.md moi).
- B1 sweep all-inherit floor-song: 15 dong/5 file + 3 cung-lop lot-pattern + 5 dong H23-stale o 3 command-file (H1 bat) -> VA HET; spawn-model-audit Section-D flush H8->S124 + effort-check (fault-inject 2-chieu PASS, 14 duong-gia -> 0); governance-detectors C3 seed +worker-tier-pin; re-grep 0 floor-song; suite 45 = baseline 0-new.
- vocab-alias-map §3 canonical flip -> worker-tier-pin (all-inherit/two-tier = 2 alias cu); STATUS:6 FROZEN-diary label (het H24-3 lag); STATUS:25 + agents/README:5 sync.
- Email hub h23-precedence-test-result stamp 945cf3ad (reviewer-gate FAIL bat 3 loi truoc stamp -> gotcha #83 aggregate-cumulative) + _index cung luot.
- B3/B4 owner-gated: soan b3-b4-owner-gated-proposals.md CHO ANH (mega-line retier + budget:74 wording) — 0 apply.
- Closeout: harvest 4 lane + 2 monitor (H2 GATE PASS 5/5, 0/29 orphan; H1 3/4 clean + 5-spot fix) + curate moved-not-cut reviewer/inv-cb + carry disposition (DONG 2: restart-s124, h23-fable-test-postrestart; MO re-stamp 9) + H24 counter=5 + pilot H22 D5 (pause-roundtrip + hook-fire re-verified; cross-model PENDING).
- Governance/tooling-only: 0 prod-code, 0 mig, test 509 GIU, bundle FROZEN.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

scripts/governance-detectors.ps1

@@ -432,8 +432,11 @@ Write-Section 'C3 - vocab-fork'
 $aliasSets = @(
     @('wave-folder', 'run-trace'),
     @($VN_DUTRU_PRO, $VN_NGANSACH_PRO),
-    @('two-tier', 'all-inherit')
+    @('two-tier', 'all-inherit', 'worker-tier-pin')
 )
+# NOTE (S126, 2026-07-16): 'worker-tier-pin' added as 3rd variant - owner-decision S124 made it
+# the canonical term (vocab-alias-map.md section 3); the B1 sweep found the fork all-inherit-vs-
+# worker-tier-pin LIVE in 5 governance files (patched S126) and this seed could not see it.
 # NOTE (Harness-16, S93): "memory-fidelity" is INTENTIONALLY NOT seeded here. It is
 # NOT a vocab-fork (one concept / two names) -- it is two DISTINCT concepts sharing a
 # word: H6.7 "memoryDelta-routing-fidelity" (delta lands in the right agent-memory)

scripts/spawn-model-audit.ps1

@@ -30,9 +30,13 @@
          runs. It cannot pre-verify what the NEXT spawn will resolve to. A clean report does NOT
          promise the next spawn stays on the expected version. This is exactly why H23 section 2(4)
          makes it informational, not enforcement.
-    (ii) PRECEDENCE vs a frontmatter HARD-PIN (the hub form: 17/17 'model: claude-opus-4-8').
-         UNTESTED at SE and UNTESTABLE here: SE has 0 hard-pinned agents (12/12 'model: inherit',
-         owner order S66/H8). Only the inherit case has evidence - see the next block.
+    (ii) PRECEDENCE vs a frontmatter HARD-PIN - MEASURED at SE 2026-07-16 (S126), after S124
+         re-pinned 14/14 agents to 'model: opus' (ALIAS pin): spawn param 'fable' on agentType
+         reviewer (pin 'opus') RESOLVED claude-fable-5 (run wf_f960dae2-fa0, 3 records); a
+         control lane with NO override resolved claude-opus-4-8 under lead=Fable => the pin
+         itself works at runtime AND the spawn param BEATS it. Caveats: alias-pin only (the
+         hub form pins FULL-ID 'claude-opus-4-8' 17/17 - NOT covered by this measurement);
+         retrospective - says nothing about future spawns. Older 'inherit' evidence: below.
     (iii)Whether the spawn param REJECTS a full-id (fix #8a). Per harness docs the param is an enum
          alias (sonnet|opus|haiku|fable); passing a full-id has NEVER been tried. Not tried here.
     (iv) Documentation coverage. This measures EXPLICITNESS IN TRANSCRIPTS only. A clean report must
@@ -52,6 +56,8 @@
     HONEST LIMIT: this does not prove 'inherit' LOSES a precedence contest. 'inherit' may simply mean
     "defer to the caller", in which case there is no contest at all. Both readings give the same
     operative answer (the param takes effect); this data cannot separate them, and does not try.
+    UPDATE 2026-07-16 (S126): the hard-pin contest HAS now been measured - see NOT-MEASURABLE (ii)
+    above, which flipped to MEASURED. Param 'fable' beat frontmatter 'model: opus' on a real lane.
 
 .PARAMETER RepoRoot
   Repo root. Default = 2 levels up from this script (scripts/ -> repo root).
@@ -389,9 +395,10 @@ else {
 }
 
 # ===========================================================================
-# SECTION D - H8 all-inherit: agent frontmatter
+# SECTION D - S124 worker-tier-pin: agent frontmatter (model: opus + effort: max)
+# (was: H8 all-inherit - superseded by owner-decision S124 2026-07-15, adap erratum-h8)
 # ===========================================================================
-Write-Section 'D - frontmatter all-inherit (H8)'
+Write-Section 'D - frontmatter worker-tier-pin (S124)'
 
 # GLOB, never a hardcoded roster size: the roster moves 12 -> 14 at W2, and a hardcoded count
 # would self-age into a false FAIL the moment it lands (fix #9(h) / R2-M3).
@@ -403,13 +410,16 @@ if (-not (Test-Path $agentDir)) {
 else {
     $agentFiles = @(Get-ChildItem -Path $agentDir -Filter *.md -File -ErrorAction SilentlyContinue |
             Where-Object { $_.Name -ne 'README.md' })
-    $inheritCount = 0
+    $pinCount = 0
+    $effortOkCount = 0
 
     foreach ($a in $agentFiles) {
         $lines = @(Get-Content -Path $a.FullName -Encoding UTF8)
         # frontmatter = the block between the first '---' and the next '---'
         $modelVal = $null
         $modelLine = 0
+        $effortVal = $null
+        $effortLine = 0
         $inFm = $false
         for ($i = 0; $i -lt $lines.Count; $i++) {
             $t = $lines[$i].Trim()
@@ -421,30 +431,49 @@ else {
                 $modelVal = $Matches[1].Trim()
                 $modelLine = $i + 1
             }
+            if ($inFm -and $t -match '^effort\s*:\s*(.+?)\s*$') {
+                $effortVal = $Matches[1].Trim()
+                $effortLine = $i + 1
+            }
         }
 
         if ($null -eq $modelVal) {
             Write-Flag 'MED' (Rel $a.FullName) `
-                'no "model:" key in frontmatter - the lane falls back to a harness default, NOT inherit' `
-                'add "model: inherit" to the agent frontmatter (H8 all-inherit)'
+                'no "model:" key in frontmatter - the lane falls back to a harness default, NOT the worker-tier pin' `
+                'add "model: opus" to the agent frontmatter (S124 worker-tier-pin)'
         }
-        elseif ($modelVal -cne 'inherit') {
+        elseif ($modelVal -cne 'opus') {
             Write-Flag 'MED' ("{0}:{1}" -f (Rel $a.FullName), $modelLine) `
-                ("frontmatter pins 'model: {0}' - H8 says every agent inherits the lead" -f $modelVal) `
-                'set "model: inherit", or have the owner ratify the pin and record it in STATUS.md Sub-agents'
+                ("frontmatter has 'model: {0}' - S124 worker-tier-pin expects every agent pinned 'model: opus'" -f $modelVal) `
+                'set "model: opus", or have the owner ratify the change and record it in STATUS.md Sub-agents'
         }
         else {
-            $inheritCount++
+            $pinCount++
+        }
+
+        if ($null -eq $effortVal) {
+            Write-Flag 'MED' (Rel $a.FullName) `
+                'no "effort:" key in frontmatter - S124 worker-tier-pin expects "effort: max" alongside the model pin' `
+                'add "effort: max" to the agent frontmatter (S124 worker-tier-pin)'
+        }
+        elseif ($effortVal -cne 'max') {
+            Write-Flag 'MED' ("{0}:{1}" -f (Rel $a.FullName), $effortLine) `
+                ("frontmatter has 'effort: {0}' - S124 worker-tier-pin expects 'effort: max'" -f $effortVal) `
+                'set "effort: max", or have the owner ratify the change and record it in STATUS.md Sub-agents'
+        }
+        else {
+            $effortOkCount++
         }
     }
 
     # verdict COMPUTED from the counts, never a pasted label (W0.4 lesson 2)
-    Write-Host ("  frontmatter: {0}/{1} agent .md carry 'model: inherit'" -f $inheritCount, $agentFiles.Count)
+    Write-Host ("  frontmatter: {0}/{1} agent .md carry 'model: opus' and {2}/{1} carry 'effort: max' (S124 worker-tier-pin)" -f `
+            $pinCount, $agentFiles.Count, $effortOkCount)
     if ($agentFiles.Count -eq 0) {
         Write-Host '  [NO-SCOPE] 0 agent .md found - a clean result here would be VACUOUS.' -ForegroundColor Yellow
     }
-    elseif ($inheritCount -eq $agentFiles.Count) {
-        Write-Host ("  [OK] all-inherit holds ({0}/{0})" -f $agentFiles.Count) -ForegroundColor Green
+    elseif (($pinCount -eq $agentFiles.Count) -and ($effortOkCount -eq $agentFiles.Count)) {
+        Write-Host ("  [OK] worker-tier-pin holds ({0}/{0} model + {0}/{0} effort)" -f $agentFiles.Count) -ForegroundColor Green
     }
 }
 
@@ -457,8 +486,9 @@ Write-Host '  1. SPOT-CHECK BY CONVENTION, NOT ENFORCEMENT. Nothing here blocks
 Write-Host '  2. CURRENT SESSION ONLY. Older runs predate the convention; scanning them yields noise, not a bug storm.'
 Write-Host '  3. NOT "the floor is applied". This measures explicitness in transcripts, never documentation coverage.'
 Write-Host '  4. RETROSPECTIVE ONLY. It reads models that ALREADY resolved; it cannot pre-verify the NEXT spawn.'
-Write-Host '  5. NOT TESTED - spawn param vs a frontmatter HARD-PIN: SE has 0 hard-pinned agents (12/12 inherit),'
-Write-Host '     so that precedence case has no evidence here. The inherit case IS measured (see header, wf_cb964f83-331).'
+Write-Host '  5. TESTED 2026-07-16 (S126) - spawn param vs frontmatter HARD-PIN: the param WON (reviewer pinned opus +'
+Write-Host '     spawn fable -> claude-fable-5; no-override control -> claude-opus-4-8 under lead=Fable). ALIAS-pin only;'
+Write-Host '     the full-id-pin form (hub) is still untested here. Inherit-case evidence: header, wf_cb964f83-331.'
 Write-Host '  6. NOT TRIED - whether the spawn param rejects a full-id (fix #8a). Docs say enum alias; never attempted.'
 
 # ---------------------------------------------------------------------------