session 2026-08-17: PENDING-142/143/144/145 + REVIEWED-122/123 executed; wake detector rebuilt; 16 trackers stamped; 39 files' frontmatter repaired

Session record, ledger, index rotation and KG appends for the day the
instruments were audited and lost.

Filed: PENDING-142 (open/closed criterion answers an adjacent question) with
three addenda, PENDING-143 (carrier restoring PENDING-121 by hand), PENDING-144
(script-resident substrate claims are checked by nothing), PENDING-145 (a ruling
claims a NUMBER, not a record — PENDING-131's addenda suppressed on arrival, and
the unbuilt fence has never appeared in the open list).

Executed: REVIEWED-122 conds. 6/7/9-as-amended/11 and REVIEWED-123 conds. 1/2/3.
N-now recorded at 60/84 transcripts. Ladder frozen generally; 4 rows queued in
PENDING-141's owed-entries list, two of them earned today (mtime-is-not-content-age;
verify a bulk edit against the pre-change state from git). OWED-4 flagged as a
REWORDING to merge on lift rather than append beside.

Index rotated: prior Active Session demoted verbatim to MEMORY-reference.md,
new one promoted. MEMORY.md 20,242 bytes (83% of the 24.4 KB read limit) —
under budget but the restructuring task remains owed.

KG: 7 lines — 4 drift-patterns, 3 preventions, including the DEGRADED banner
catching its own author's regression and a confound filed against the executor's
own favourable evidence.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013Y6t6qx7cpaCu5xGdD36u4

claude/memory/knowledge-graph.jsonl

@@ -662,3 +662,10 @@
 {"subject": "contamination-problem.md", "predicate": "scope-limit", "object": "IS A THEORY OF ONE MISALIGNMENT FLAVOUR, USED HERE AS THE THEORY OF EXECUTOR FAILURE IN GENERAL. Against Byrnes's four-flavour taxonomy (imitative→seven-sins · human-approval→glazing · automatic-verifiers→literal-genie · LLM-judges→trickster), every mitigation in the doc — behavioural observation, explicit permission structures, indirect questioning, longitudinal analysis — is calibrated against APPROVAL-SEEKING, i.e. glazing. A crude keyword probe over the 235 banked claude-code drift-patterns classified 106 and left 129 unclassified: 86 literal-genie, 12 trickster, 8 glazing, 0 seven-sins. ⚠ The classifier is keyword-matching over prose — the exact defect PENDING-139 names — so indicative, not measured. If the skew survives a real instrument, our doctrine is ~100% anti-sycophancy while our failures are dominated by verifier-Goodhart, against which a control is simply another proxy.", "valid_from": "2026-08-17", "valid_to": null, "confidence": 0.6, "source_file": "session-2026-08-14-the-controls-tested-the-wrong-property.md", "extracted_at": "2026-08-17"}
 {"subject": "a governed record's own formatting", "predicate": "drift-pattern", "object": "CAN MAKE AN ENTRY INVISIBLE TO THE CHECK THAT GUARDS IT, WHILE THE CHECK REPORTS CLEAN. REVIEWED-121 AMENDMENT 1 was placed truncated (ended mid-A3), then re-pasted with an unclosed ```yaml fence that swallowed A3's binding rule, all of A4 and the disposition — rendering them as code and stripping their emphasis — and would have swallowed the NEXT entry appended to REVIEWED.md. Register-integrity reported clean throughout, because its subject is HEADINGS, not fences. The re-paste also indented the body 2 spaces; had it indented the HEADING, RE_HEAD (^##\\s+) would have stopped matching and the amendment would have vanished from the check entirely with no alarm. Third instance in one week of a check whose subject sits adjacent to the property that matters.", "valid_from": "2026-08-17", "valid_to": null, "confidence": 1.0, "source_file": "session-2026-08-14-the-controls-tested-the-wrong-property.md", "extracted_at": "2026-08-17"}
 {"subject": "an AUTHORIZED-and-unblocked action", "predicate": "drift-pattern", "object": "CAN STILL BE THE WRONG ACT, AND THE MEMORY INDEX WAS PUSHING TOWARD IT. The 41 S2 ladder rows are authorized (2026-07-19) and MEMORY.md described them as 'needing execution not a ruling' and 'unblocked' — both true as to authorization. But appending them triples the verification ladder from 20 entries WHILE a pre-registered trial measures whether the ladder is reached (baseline 14%, graded at 84 transcripts), and ladder SIZE is an uncontrolled variable in that design. A session doing exactly the right procedural thing would have confounded the only check behind REVIEWED-95's causal claim. Filed PENDING-141, recommendation (a) HOLD. ⚠ The index line was amended in the SAME act as the filing — a finding that leaves the misleading line standing is a note, not a finding. Authorization answers 'may I', never 'should I now'.", "valid_from": "2026-08-17", "valid_to": null, "confidence": 1.0, "source_file": "session-2026-08-14-the-controls-tested-the-wrong-property.md", "extracted_at": "2026-08-17"}
+{"subject": "claude-code", "predicate": "drift-pattern", "object": "THE FIX REPRODUCED THE DEFECT IT WAS FIXING, IN THE TOOL WRITTEN TO CLEAN UP AFTER IT. `strip_frontmatter`'s non-greedy `^---\\n.*?\\n---\\n` was diagnosed as matching the STRAY frontmatter block on 39 damaged files; one hour later the stamping script written to mark superseded trackers used the identical pattern, matched the stray block on 3 of them, and orphaned the real frontmatter into the body. The post-stamp check reported `malformed: none` because its subject was 'does the file begin with frontmatter then a banner' (true) while the claim was 'the stamp preserved the record's keys' (false). A diagnosis held in working memory does not transfer to the next tool written unless the pattern itself is searched for.", "valid_from": "2026-08-17", "valid_to": null, "confidence": 1.0, "source_file": "session-2026-08-17-the-instruments-audited-themselves-and-lost.md", "extracted_at": "2026-08-17"}
+{"subject": "a file's timestamp", "predicate": "drift-pattern", "object": "IS NOT ITS CONTENT'S AGE, AND NEITHER IS GIT'S LAST-COMMIT. Three instances 2026-08-17: 191 wrap records dated by mtime, so a CODA appended three days later made the 08-14 session look unwrapped; 61 trackers reported at exactly 72.3 days by mtime AND by git-date, both reset by 3f9a89b, a 283-file normalization sweep — three successive staleness estimates were wrong before the fourth excluded it; and repairing 20 April-May wrap records moved their mtimes to today and promoted an April session to `Last wrap`, losing the pulling thread and the open question. The third was caused by the fix for the first. Reach for the earliest signal an ordinary later act cannot move, and exclude known bulk operations before quoting an age.", "valid_from": "2026-08-17", "valid_to": null, "confidence": 1.0, "source_file": "session-2026-08-17-the-instruments-audited-themselves-and-lost.md", "extracted_at": "2026-08-17"}
+{"subject": "a ruling in REVIEWED.md", "predicate": "drift-pattern", "object": "CLAIMS A NUMBER, NOT A RECORD — so every addendum filed under that number afterwards is suppressed ON ARRIVAL. REVIEWED-115 (2026-08-10) claimed `131`; PENDING-131 ADDENDUM 4 was filed 2026-08-13 and has never appeared in the open list, while awaiting steward direction. PENDING-131 (c) is the unbuilt fence — the pulling thread of every session since 08-10, made a CONDITION by REVIEWED-121 — and the instrument whose job is reporting what awaits authorization has been silent about it for a week. The work survived only because MEMORY.md and the session records carried it by hand. The same item fails the other way too: REVIEWED-116's header `PENDING-131/132/133/134` parses to one token matching no id, so a four-item ruling suppresses nothing. Filed PENDING-145.", "valid_from": "2026-08-17", "valid_to": null, "confidence": 1.0, "source_file": "session-2026-08-17-the-instruments-audited-themselves-and-lost.md", "extracted_at": "2026-08-17"}
+{"subject": "a size guard on a scanner", "predicate": "drift-pattern", "object": "BECOMES A SILENT BLIND SPOT THE MOMENT THE FILE IT GUARDS IS THE ONE THAT MATTERS. governance-drift-check.py skipped any file over 400 KB; PENDING.md reached 546,944 bytes, so every structured DEFERRED-DECISION block in the governance register went unread — by the checker built to stop deferred conditions being silently missed. Compounded because the prose-deferral loop has NO guard, so PENDING.md's prose count kept appearing in the report and made the file look examined. Found only by placing a block under REVIEWED-123 cond. 2 and noticing the tracked count did not move. Fixed: governance files exempt, and a skip is now reported by name with byte counts as 'could not assess'.", "valid_from": "2026-08-17", "valid_to": null, "confidence": 1.0, "source_file": "session-2026-08-17-the-instruments-audited-themselves-and-lost.md", "extracted_at": "2026-08-17"}
+{"subject": "the honest-degradation banner", "predicate": "prevention", "object": "CAUGHT ITS OWN AUTHOR'S REGRESSION IN-SESSION. Repairing 39 files' frontmatter moved 20 April-May wrap records' mtimes to today; sec_pause picked `newest wrap` by mtime and promoted an April session to Last wrap, losing both the pulling thread and the open question. Nothing in the repair's own verification would have caught it — the repair's checks were about frontmatter, and they all passed. The DEGRADED section fired instead ('no PULLING THREAD anchor in the last wrap'), which is Constitutional Constraint 4 built in 2026-07-28 for a different reason and paying out here. A mechanism that reports its own limits caught a defect no purpose-built control was pointed at.", "valid_from": "2026-08-17", "valid_to": null, "confidence": 1.0, "source_file": "session-2026-08-17-the-instruments-audited-themselves-and-lost.md", "extracted_at": "2026-08-17"}
+{"subject": "reading the PLACED record instead of the relayed message", "predicate": "prevention", "object": "FIFTH CONSECUTIVE FIRING, AND IT CHANGED THE WORK BOTH TIMES ON 2026-08-17. The relayed REVIEWED-122 gave the ruling; the placed entry carried five conditions, two SEVERED legs authorized ahead of the mechanism, and a binding ORDER (answer key first, alone, hash recorded, before any implementation). The relayed REVIEWED-123 read as a ladder hold for the S2 batch; the placed entry ruled the freeze GENERAL — from any source, whatever its authorization — which is what made queueing today's two earned entries to the owed list correct rather than a violation.", "valid_from": "2026-08-17", "valid_to": null, "confidence": 1.0, "source_file": "session-2026-08-17-the-instruments-audited-themselves-and-lost.md", "extracted_at": "2026-08-17"}
+{"subject": "declaring a confound against one's own favourable evidence", "predicate": "prevention", "object": "STOPPED A FLATTERING n=2 FROM STANDING UNQUALIFIED. The jurist offered a discriminator for real-vs-manufactured authorization boundaries, with the executor's 2026-08-17 refusal as its POSITIVE instance. The two cases differ in a variable the discriminator does not name: on 08-17 PENDING-141 was in MEMORY.md's Active Session block, bold, flagged 'do not execute', and read at that session's wake; on 08-01 no equivalent prompt existed. So the positive instance may record an INDEX that named the instrument rather than an executor that found it — the discriminator would then measure the memory layer while appearing to measure judgment. Filed by the party the evidence flatters, into the record, before anyone asked.", "valid_from": "2026-08-17", "valid_to": null, "confidence": 0.9, "source_file": "session-2026-08-17-the-instruments-audited-themselves-and-lost.md", "extracted_at": "2026-08-17"}