session 2026-07-27 evening: PENDING-76/77/78 placed; governance drift-check built + wired into /wake-up

PENDING-76 remanded by jurist — required count returned 0 of 11 (the package's own
IV.2 refinement proved its target category empty); executor recommends withdrawal.
PENDING-77 (5 structural defects) and PENDING-78 (.app preferences) released by the
ruling from needing it. Drift check reports contradicted state claims at every wake
and corrects nothing — detection needs no authorization, correction does.

MEMORY.md compacted 20.5KB -> 17.1KB (budget hook); prior Active Session demoted to
MEMORY-reference.md. CLAUDE.md and REVIEWED.md untouched.

Co-Authored-By: Claude <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Xefg5EXwcpd9RMAr63dWrD

claude/memory/knowledge-graph.jsonl

@@ -456,3 +456,9 @@
 {"subject": "claude-code", "predicate": "drift-pattern-good-direction", "object": "a-named-dependency-can-DISSOLVE rather than get solved — I flagged 'the attestation binds the landed sha but the gate runs at graduation' as blocking. Reading the apply leg showed graduation is a PURE MOVE (git mv/rename, no content transformation), so candidate bytes ARE landed bytes and the seam does not exist. Steward: 'the better kind of dependency resolution — it dissolved rather than got solved, and the reason is checkable in the code.' Reusable: before designing around a dependency, read whether it is real.", "valid_from": "2026-07-27", "valid_to": null, "confidence": 1.0, "source_file": "session-2026-07-25-to-27-two-supersessions-and-the-sweep-that-caught-its-own-fix.md", "extracted_at": "2026-07-27"}
 {"subject": "chamber-library", "predicate": "spec-version", "object": "v2.8.0 OPERATIVE (2026-07-27) — v2.7.0 REVIEWED-75 kind-scoping the verification criterion (method declared for the work's KIND; deterministic/scan split preserved; anti-bypass guard bound to the property); v2.8.0 REVIEWED-73 voice-purity as the engine-consumable bar (exhaustive partition · closed-vocab role · chamber-side · attest-never-default). v2.7.0 and v2.6.0 frozen.", "valid_from": "2026-07-27", "valid_to": null, "confidence": 1.0, "source_file": "session-2026-07-25-to-27-two-supersessions-and-the-sweep-that-caught-its-own-fix.md", "extracted_at": "2026-07-27"}
 {"subject": "chamber-library", "predicate": "open-defect", "object": "musil-mwq-tome-1 served to the engine with section boundaries shifted by 6 lines — canonical drifted 5032->5038; body truncated early, 6 lines mislabelled apparatus, 6 unassigned at the tail. Found by REVIEWED-73's exhaustive-partition rule on first application; HELD for re-derivation by a reading pass, NOT migrated. Also: REVIEWED-73's cited instance (5 standing FAILED ARC rows) was WRONG — all five ARC bindings verify clean.", "valid_from": "2026-07-27", "valid_to": null, "confidence": 1.0, "source_file": "session-2026-07-25-to-27-two-supersessions-and-the-sweep-that-caught-its-own-fix.md", "extracted_at": "2026-07-27"}
+{"subject": "claude-code", "predicate": "drift-pattern", "object": "relay-a-subagent-claim-as-established — twice in one session I passed a subagent's finding to the steward as fact and had to retract: 'mempalace fails silently, exit 0' (it exits 1) and then the corrected version (the typography path actually HANGS, exit 124). Both rested on `cmd | head -N; echo $?`, which reports head's status, not the command's. A subagent's confident finding is a claim to verify, not a result to forward — the same self-assessment discipline CLAUDE.md applies to code, applied to agents.", "valid_from": "2026-07-27", "valid_to": null, "confidence": 1.0, "source_file": "session-2026-07-27-evening-governance-currency-and-the-amendment-that-killed-itself.md", "extracted_at": "2026-07-27"}
+{"subject": "claude-code", "predicate": "drift-pattern", "object": "a-measurement-that-includes-the-measurer — four instrument failures in one day, all the same family: piped exit codes (x3, reporting head's status), a `ps | grep mempalace` whose count included the shell running the grep, and a `find -maxdepth 4` whose positive control sat at depth 5. Every one produced a confident wrong number. RULE (jurist-ratified 2026-07-27, Q2): a negative command result requires a positive control in the same invocation — an absence proves nothing until the instrument is shown capable of detecting presence.", "valid_from": "2026-07-27", "valid_to": null, "confidence": 1.0, "source_file": "session-2026-07-27-evening-governance-currency-and-the-amendment-that-killed-itself.md", "extracted_at": "2026-07-27"}
+{"subject": "claude-code", "predicate": "drift-pattern", "object": "dont-run-your-own-refinement-back-across-your-own-census — I authored a jurist package whose best part (IV.2, splitting machine-verifiable from steward-held state) was ALSO fatal to the proposal containing it, and never re-tested the Part II census against it. The jurist ran that check in one pass: required count 0 of 11. Reusable: when a draft gains a sharper test late, re-run every earlier claim through it before filing — the new test is most dangerous to the argument that produced it.", "valid_from": "2026-07-27", "valid_to": null, "confidence": 1.0, "source_file": "session-2026-07-27-evening-governance-currency-and-the-amendment-that-killed-itself.md", "extracted_at": "2026-07-27"}
+{"subject": "claude-code", "predicate": "drift-pattern", "object": "an-eval-designed-run-and-read-by-the-proposing-party — I built a 3-arm eval to test my own proposed CLAUDE.md block and it returned the most favourable available finding. I flagged n=1 (precision); the jurist flagged CONSTRUCT VALIDITY — 3 tasks contain no tail, so guardrail redundancy was never measurable at all. Contamination operating structurally, not as bad faith. Reusable: when the eval supports the evaluator's proposal, the question is not 'is the sample big enough' but 'can this design measure the thing at all'.", "valid_from": "2026-07-27", "valid_to": null, "confidence": 1.0, "source_file": "session-2026-07-27-evening-governance-currency-and-the-amendment-that-killed-itself.md", "extracted_at": "2026-07-27"}
+{"subject": "governance-drift-check", "predicate": "is", "object": "a wake-time detector at ~/dotfiles/scripts/governance-drift-check.py, wired into /wake-up section 2.c. Reports state claims in ~/CLAUDE.md the substrate contradicts (unresolvable paths, unresolvable tool names, unconfigured hooks claimed to fire, expired date horizons, structural damage). 0.17s; every check carries a same-run positive control. Detects, never corrects — detection needs no authorization, correction does. Reported 9 findings at build time 2026-07-27.", "valid_from": "2026-07-27", "valid_to": null, "confidence": 1.0, "source_file": "session-2026-07-27-evening-governance-currency-and-the-amendment-that-killed-itself.md", "extracted_at": "2026-07-27"}
+{"subject": "typography-palace", "predicate": "has-no-live-index", "object": "0 active segment directories, 6 quarantined; collection 865f3806 has had none since 2026-06-02. Cannot serve a query — a real search hangs to timeout (exit 124). Source texts were ingested FROM chamber-library and all survive there (steward-confirmed); palace-native content is 10 entities / 7 triples. Retirement = export 17 KG rows to JSONL, delete 480MB. reference-typography-palace-cli.md documents an instrument that no longer exists.", "valid_from": "2026-07-27", "valid_to": null, "confidence": 1.0, "source_file": "session-2026-07-27-evening-governance-currency-and-the-amendment-that-killed-itself.md", "extracted_at": "2026-07-27"}