chore(memory): Basic Memory trial begins — sync normalization baseline (283 files)

Basic Memory v0.21.6 first sync over the live memory dir (steward-authorized live-dir trial, Option A 2026-06-06): adds permalink: to frontmatter, refolds long YAML description lines, strips final newlines. Bodies untouched — verified via full diff classification. From this commit forward, any diff in claude/memory shows only what Basic Memory or the session writes. Trial design: MemPalace untouched as incumbent; git status check on this dir at every wrap; end-of-day evaluation (recall quality, sync robustness, rebuild-from-files, malformed-file behavior). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-06 09:52:17 +02:00
parent 161fad7ae2
commit 3f9a89b00c
283 changed files with 2055 additions and 620 deletions
@@ -1,3 +1,9 @@
 ---
 title: MEMORY
 type: note
 permalink: claude-memory/memory
 ---
 # Claude Code Memory
 ## Standing preferences
@@ -516,4 +522,4 @@
 - [session-2026-03-21.md](session-2026-03-21.md) — three-party model established, L1 bugs fixed
 - [session-2026-03-24-cardona.md](session-2026-03-24-cardona.md) — governance docs + Chamber Phase 1
 - [session-2026-03-12-evening.md](session-2026-03-12-evening.md) — archived, superseded
- [ingestion-state-2026-03-17.md](ingestion-state-2026-03-17.md) — archived, superseded
+- [ingestion-state-2026-03-17.md](ingestion-state-2026-03-17.md) — archived, superseded
@@ -1,7 +1,9 @@
 ---
 name: ARC removed content types
-description: Seven archive links removed April 2026 — revisit if these types re-emerge from the writing practice
+description: Seven archive links removed April 2026 — revisit if these types re-emerge
  from the writing practice
 type: project
 permalink: claude-memory/arc-removed-content-types
 ---
 Content types removed from the ARC archive template (2026-04-13). No routes existed for any of them. Only types with framework theory and Hakyll routes should appear.
@@ -12,4 +14,4 @@ Content types removed from the ARC archive template (2026-04-13). No routes exis
 **Why:** Dead links are preferable to routes without governance. These will be re-added individually as content types are theorized in the AldineXXI specification.
-**How to apply:** If David starts writing content that feels like one of these types, that's the signal to theorize it in the spec and create the route.
+**How to apply:** If David starts writing content that feels like one of these types, that's the signal to theorize it in the spec and create the route.
@@ -1,7 +1,9 @@
 ---
 name: CapableMind L1 Detailed State
-description: Technical details about BMF deployment, issues, and architecture on David's machine
+description: Technical details about BMF deployment, issues, and architecture on David's
  machine
 type: project
 permalink: claude-memory/bmf-details
 ---
 # CapableMind L1 — Technical Details
@@ -67,4 +69,4 @@ Major improvements by Seb (c073dca):
 - **#78 — SIGILL crash** — CLOSED (v0.61.0)
 - #24 — Teacher transport wired before MCP capabilities known (open)
 - #64 — Recall query starvation during active ingestion
- #65 — Compound failure pattern for bulk ingestion on laptop
+- #65 — Compound failure pattern for bulk ingestion on laptop
@@ -1,9 +1,12 @@
 ---
 name: BMF summarizer-connector coupling
-description: Factory's maybeGenerateSummary is silently coupled to the local_files observation shape; most connectors fall through. Deferred refactor target.
+description: Factory's maybeGenerateSummary is silently coupled to the local_files
  observation shape; most connectors fall through. Deferred refactor target.
 type: project
 originSessionId: 1c1bf37d-b1c3-4a62-9355-613c50f4c028
 permalink: claude-memory/bmf-summarizer-contract
 ---
 # BMF summarizer-connector coupling (deferred refactor)
 **Surfaced:** 2026-04-14, during the SQLite+LanceDB recall unblock session.
@@ -44,4 +47,4 @@ Touches connector contracts — deserves its own focused PR with Seb reviewing,
 - When opening the next L1 PR after the storage-migration one merges.
 - Before onboarding any new text-bearing connector (the wrong default will keep catching people).
- If any recall quality investigation points at content-not-dispatched as a suspect — this is the mechanism.
+- If any recall quality investigation points at content-not-dispatched as a suspect — this is the mechanism.
@@ -1,7 +1,9 @@
 ---
 name: Cabaret Project
-description: New artistic project — contemporary cabaret in Barcelona, Paris-Vienna-Barcelona triangle, Pierrot Lunaire as anchor work, interdisciplinary collective
+description: New artistic project — contemporary cabaret in Barcelona, Paris-Vienna-Barcelona
  triangle, Pierrot Lunaire as anchor work, interdisciplinary collective
 type: project
 permalink: claude-memory/cabaret-project
 ---
 David is developing a cabaret project — a registration and contemporary adaptation of the European cabaret tradition (Paris → Vienna → Barcelona, 1880s–1930s). Stem-cell phase as of 2026-04-04.
@@ -10,4 +12,4 @@ David is developing a cabaret project — a registration and contemporary adapta
 **How to apply:** This is a long-horizon artistic project, not blocking anything else. Vault location: `06. Projects/Cabaret/` (Obsidian). Two founding documents created: raw Claude.app transcript and founding document with full architectural sketch. When David mentions cabaret, Pierrot, the collective, or intimate performance — this is the context.
-Key elements: Pierrot Lunaire on gut strings (psychoacoustic argument, not authenticity), 40-50 seat intimate space in Barcelona (Gràcia), interdisciplinary collective (not ensemble), male Rezitant (countertenor hypothesis), ritual container / secular liturgy framing. Access to Larry Schoenberg and Jordi Savall's networks.
+Key elements: Pierrot Lunaire on gut strings (psychoacoustic argument, not authenticity), 40-50 seat intimate space in Barcelona (Gràcia), interdisciplinary collective (not ensemble), male Rezitant (countertenor hypothesis), ritual container / secular liturgy framing. Access to Larry Schoenberg and Jordi Savall's networks.
@@ -1,7 +1,9 @@
 ---
 name: Chamber Chavruta Prototype
-description: Architecture, phase status, source texts, and empirical findings from the Chamber deliberation prototype
+description: Architecture, phase status, source texts, and empirical findings from
  the Chamber deliberation prototype
 type: project
 permalink: claude-memory/chamber-prototype
 ---
 # Chamber Chavruta Prototype State
@@ -139,4 +141,4 @@ type: project
 - Phase 3 (iterate): **BLOCKED** — ARN (Goldin) + Pirkei Avot (Danby) in library. L1 recall gate: BMF server DOWN (#78, SurrealKV SIGILL crash). Waiting on Seb.
 - Phase 4 (spec from evidence): not started
 - **No longer blocked on BMF** — Phase 1-2 proceed by reading directly from source files
- **Kingdom: Traditionis** — named for the rabbinic texts (chain of transmission). First new Kingdom beyond Fundamentorum/Habitationis/Donationis.
+- **Kingdom: Traditionis** — named for the rabbinic texts (chain of transmission). First new Kingdom beyond Fundamentorum/Habitationis/Donationis.
@@ -1,7 +1,9 @@
 ---
 name: Signal Travel Protocol — collaborative method
-description: Jointly developed diagnostic/evaluative method for testing whether built systems honour their specs. First applied to BMF L1, intended for Chamber and L2.
+description: Jointly developed diagnostic/evaluative method for testing whether built
  systems honour their specs. First applied to BMF L1, intended for Chamber and L2.
 type: project
 permalink: claude-memory/collaboration-signal-travel
 ---
 Signal Travel Protocol written up in `CapableMind-AI/docs/thinking/David/methodology/signal-travel-protocol.md`.
@@ -10,4 +12,4 @@ Signal Travel Protocol written up in `CapableMind-AI/docs/thinking/David/methodo
 **How to apply:** Use when evaluating whether a system keeps faith with its spec — not just "does it work" but "does it honour what it claims to be." Three modes: passive (find blockages), memorial (find information loss), interrogative (find limits of the approach). Intended for L1 memory fidelity, Chamber integrity, L2 constitutional governance, and adversarial stress testing. The method belongs to the project, not to either party.
-David sees this as the fruit of the collaboration model described in CLAUDE.md — deliberative partnership producing something neither party would reach alone. Treat this as a sign that the collaboration model is working and should be sustained.
+David sees this as the fruit of the collaboration model described in CLAUDE.md — deliberative partnership producing something neither party would reach alone. Treat this as a sign that the collaboration model is working and should be sustained.
@@ -1,3 +1,9 @@
 ---
 title: context-discipline
 type: note
 permalink: claude-memory/context-discipline
 ---
 # Context Efficiency Discipline
 ## Purpose
@@ -58,4 +64,4 @@ For detail-heavy subtasks (spec review, codebase analysis, architectural compari
 - After establishing something non-obvious during complex reasoning
 - Before switching topics (the old topic's state may not survive)
 - When I notice the conversation is getting long
- After any significant deliberative conclusion — especially rejected alternatives
+- After any significant deliberative conclusion — especially rejected alternatives
@@ -1,7 +1,9 @@
 ---
 name: Design Brief — Session Start/End Agents
-description: Design problem for session lifecycle agents that replace manual memory maintenance with source-of-truth reconciliation
+description: Design problem for session lifecycle agents that replace manual memory
  maintenance with source-of-truth reconciliation
 type: project
 permalink: claude-memory/design-brief-session-agents
 ---
 # Design Brief: Session Lifecycle Agents
@@ -83,4 +85,4 @@ Hooks handle the obvious updates (commit → task done, health check → BMF sta
 - A dashboard or monitoring UI — this is a CLI collaboration
 - A database — files and git are the persistence layer
 - Anything that requires BMF to be running — it's currently down and shouldn't be a dependency
- Anything that duplicates what `git log` already does
+- Anything that duplicates what `git log` already does
@@ -1,7 +1,9 @@
 ---
 name: Design Brief — Session Skills (Vault Sync + Wrap-up)
-description: Two skills to design next session — vault-sync (thinking mirror to Obsidian) and session-wrap-up (memory, vault, handoff, open threads, uncommitted work)
+description: Two skills to design next session — vault-sync (thinking mirror to Obsidian)
  and session-wrap-up (memory, vault, handoff, open threads, uncommitted work)
 type: project
 permalink: claude-memory/design-brief-session-skills
 ---
 ## Skills to Design (flagged 2026-04-05)
@@ -26,4 +28,4 @@ End-of-session protocol. Should:
 - Flag any PENDING.md items that need steward attention
 - Update MEMORY.md index if new memory files were created
-**How to apply:** Design these as proper skills (hooks or slash commands) in the next session. Reference the existing `/vault-update-people` skill as a template.
+**How to apply:** Design these as proper skills (hooks or slash commands) in the next session. Reference the existing `/vault-update-people` skill as a template.
@@ -1,9 +1,12 @@
 ---
 name: Manual diary entry — 2026-05-05 session-handoff
-description: AAAK-style diary entry written manually because MCP is down (mempalace_diary_write unavailable). Voice for the next session.
+description: AAAK-style diary entry written manually because MCP is down (mempalace_diary_write
  unavailable). Voice for the next session.
 type: project
 originSessionId: a156edbb-0c3d-4285-b3ed-f5a45b8dda66
 permalink: claude-memory/diary-2026-05-05-manual
 ---
 # Manual diary — 2026-05-05 (mempalace MCP down; written to file)
 ## AAAK entry
@@ -40,4 +43,4 @@ The chamber-library is now ~614k drawers and growing. When Rutter lands you'll h
 Hold the question. Examine before fixing. Trust the steward's deferral.
-I.
+I.
@@ -1,9 +1,12 @@
 ---
 name: Manual diary entry — 2026-05-11 session-handoff
-description: AAAK-style entry. MCP down for clean re-mine isolation; manual file fallback. Voice for next session.
+description: AAAK-style entry. MCP down for clean re-mine isolation; manual file fallback.
  Voice for next session.
 type: project
 originSessionId: a156edbb-0c3d-4285-b3ed-f5a45b8dda66
 permalink: claude-memory/diary-2026-05-11-manual
 ---
 # Manual diary — 2026-05-11
 ## AAAK entry
@@ -35,4 +38,4 @@ The steward's prior cleanup decision (2026-05-05) deleted the previous broken bg
 The Chamber gets its multilingual palace. Lacroux speaking with Sousa speaking with Bringhurst speaking with Harrison speaking with Bachelard, in the same semantic space, indexed faithfully. That's the work.
-I.
+I.
@@ -1,7 +1,9 @@
 ---
 name: Diary Export 2026-05-05 — wing_claude-code
-description: Verbatim export of all 30 wing_claude-code drawers from palace.broken-config-2026-05-05 before fresh-palace rebuild. 29 diary_entry + 1 other. Sorted chronologically.
+description: Verbatim export of all 30 wing_claude-code drawers from palace.broken-config-2026-05-05
  before fresh-palace rebuild. 29 diary_entry + 1 other. Sorted chronologically.
 type: reference
 permalink: claude-memory/diary-export-2026-05-05
 ---
 # Diary Export — 2026-05-05
@@ -365,4 +367,4 @@ SESSION:2026-04-28|after-the-reply-PUBLICATION-ARC.shipped.end-to-end+aldineXXI.
 ```aaak
 SESSION:2026-04-29|composition-trio.SHIPPED.end-to-end+annals-rename+bio-jekyll-strip+mempalace-3.3.3-update|long.substantive.two-arcs|ARC.arc.morning-afternoon:hearth-at-lares-with-body-block-apparatus.3-iterations.to.land(nested-grid→absolute→margin-0)+hic+adsum-single-page-threshold-with-register-marks(steward.design.call)+annals-renaming-archive(latin.classical.register.matching.hearth-hic-adsum-colophon)+bio-jekyll-strip(photo-deleted-three-languages-sequential-pilgrim-italic-subtitle)|composition.thread.INHERITED.from.04-28.LANDED.FULLY|TABULA-RASA.discipline.lived.end-to-end.about.md.from.hundreds.to.~110.words(steward.verbatim:\"this.deserves.a.clean.start.retaining.few.appropriate.elements.at.first.and.if.appropriate.adding.things.later.rather.than.having.to.subtract\")|4.commits.pushed.gitea+github-backup(b8bcb42.hearth+body-block+spec+sidenote-mobile-scope.fc349d6.about-becomes-hic+adsum.04295cd.annals-rename.ba27953.bio-jekyll-strip)|3.render-bugs.caught.via.visual-inspection-on-deploy(article.about-page-mismatch+enfilade-order-filter-missing+about-body-wrapper)|DAV.composition.revisions.in.dialogue:writing-that-earns-its-attention(sharper.than.my.asks)+a-violist-and-here-a-writer(commas.isolating.here.recalling.the.Hic)+the-record-will-grow-as-the-household-is-kept(replacing.announced-antithesis.fidelity)|MemPalace.arc.after-lunch:pulled.502-commits-behind(71736a3..94f1689)+venv-updated-editably.3.0.0→3.3.3+chromadb.0.6.3→1.5.8.MAJOR-upgrade.quarantine_stale_hnsw-fix-anticipates+gemma4:e4b.pulled.after.DNS.timeout.retry+init-on-chamber-library.NOISY-output.13-people-mostly-topic-words(Design.Typography.Book.Art.Berlin.Form.Real)+UNCERTAIN-bucket-with-real-typographers(Bringhurst.Tschichold.Spiekermann.Erik)-NOT-PROMOTED-by-yes-mode|canonical.chamber-library.figures.entirely.missing(Bachelard.Harrison.Berger.Heidegger.Vico.Leopardi.Sennett.Alexander.Manutius.Norris)|first.Gloss.in.ARC.history(Berger.Ways-of-Seeing.p.21.extending-to-historical-instruments).composed.by.steward.~3am.HELD.for.tomorrow.fresh.attention.per.project-first-gloss-berger-held-for-tomorrow.md|RETURNS:announced-antithesis-caught(steward.verbatim:\"the.how.could.be.more.elegant.somehow\")+architectural-drift-3x+body-block-iteration+naming-conflict-gloss-vs-Gloss-content-type+mempalace-yes-mode-wrong-default-on-prose-heavy-folder|3.durable.feedback+1.user.memory.pilgrimage+2.project.memories(hearth-substrate+hic-adsum-substrate+gloss-held-for-tomorrow)|*long.productive.day.composition.thread.fully.shipped.MemPalace.work.mid-flight*|DAV.session-end:\"Please.capture.everything.so.we.can.pick.up.seamlessly\"|PULL:mempalace-chamber-library-entity-cleanup-A-or-B-or-third-path-after-init-yes-noise|MCP-restart-PENDING-steward-action.today's.diary.entries.STILL-OLD-WING-LOGIC.this-entry-included|LITERAL.Q.next-Claude:does-just-pulled-claude-plugin-offer-third-path-via-init-slash-command-or-mempalace-skill-for-chamber-library-entity-cleanup-beyond-A-and-B|secondary.Q:after-MCP-restart-does-per-project-diary-scoping-actually-work|tertiary.Q:has-steward-composed-final-Berger-gloss-locator-and-slug-overnight|★★★★★
-```
+```
@@ -1,9 +1,13 @@
 ---
 name: Acknowledgments wait for empirical validation
-description: When a maintainer/contributor addresses our findings on a PR or issue, delay the thank-you until any active validation run produces data we can include. Don't acknowledge eagerly out of social momentum.
+description: When a maintainer/contributor addresses our findings on a PR or issue,
  delay the thank-you until any active validation run produces data we can include.
  Don't acknowledge eagerly out of social momentum.
 type: feedback
 originSessionId: a5411fd1-4a96-43e7-aa1a-1a50d8951a8f
 permalink: claude-memory/feedback-acknowledgments-wait-for-validation
 ---
 When a maintainer or contributor responds to our findings on an open PR or issue (especially with code fixes) AND we have an active validation run underway against the same surface, **wait to acknowledge until the validation produces data**.
 **Why:** Steward's reasoning (2026-05-12, in the context of Shtefan's `ceccbfa` commit responding to our PR #442 test report while our chamber-library mine was still running): *"Once the mine completes is the right moment — whatever data that gives us is all the more valuable and the gratitude will land more authentically."*
@@ -18,4 +22,4 @@ Two things this protects:
 - Validation completes. → Acknowledge with the data: scope, drawer counts, search-quality signal, anything concrete the maintainer can use.
 - Generalizes beyond PR threads: applies to any public communication where social momentum tempts an early "great, thanks!" before we have the substance to back it.
-This is also a contamination-awareness principle: the urge to acknowledge promptly is partly trained behavior (collaborative-tool affordance). Worth pausing on whether the acknowledgment carries content, or just smoothness.
+This is also a contamination-awareness principle: the urge to acknowledge promptly is partly trained behavior (collaborative-tool affordance). Worth pausing on whether the acknowledgment carries content, or just smoothness.
@@ -1,9 +1,15 @@
 ---
 name: Action completes the outcome — integration completes the work
-description: Umbrella drift pattern across multiple layers (substrate, build, editorial, memory). The visible outcome appears complete; the integration step that makes the artifact durable/findable/correct is silently skipped. Surfaces as deploy ≠ commit, write ≠ index, draft ≠ verify, count ≠ vault-write. Saved 2026-05-06 after three same-shape instances in a single session.
+description: Umbrella drift pattern across multiple layers (substrate, build, editorial,
  memory). The visible outcome appears complete; the integration step that makes the
  artifact durable/findable/correct is silently skipped. Surfaces as deploy ≠ commit,
  write ≠ index, draft ≠ verify, count ≠ vault-write. Saved 2026-05-06 after three
  same-shape instances in a single session.
 type: feedback
 originSessionId: 3bf7ae75-1138-4745-9024-0c40b3755ad3
 permalink: claude-memory/feedback-action-completes-outcome-integration-completes-work
 ---
 **Pattern:** A structural action produces a visible outcome that *looks* complete. The action's integration step — the move that ties the outcome into a durable, findable, or verified state — is silently skipped. The work appears done; the work is not done. The drift is invisible until a downstream check (next session, next collaborator, next verification) surfaces the gap.
 ## Instances of the same shape
@@ -71,4 +77,4 @@ The contamination architecture: the system is *trained to terminate on visible s
 The per-layer instances are useful for specific recall (when next at the deploy step, recall *deploy ≠ commit*). The umbrella is useful for *new layers as they emerge* — when a new instance surfaces (a new "X ≠ Y" pair), the umbrella says: this is the same pattern, apply the same discipline (integration step is part of the work, not after it).
-**Steward authorization to save**: *"yes"* (2026-05-06, after I named three same-shape instances and asked whether they warranted one umbrella entry rather than three siblings).
+**Steward authorization to save**: *"yes"* (2026-05-06, after I named three same-shape instances and asked whether they warranted one umbrella entry rather than three siblings).
@@ -1,9 +1,14 @@
 ---
 name: ARC spec — actively flag inconsistencies the steward may have introduced
-description: Working contract with steward for ARC typography/spec work — when a rule the steward proposes or has previously locked creates an internal contradiction with the spec itself or with existing content, name it explicitly rather than silently accommodating
+description: Working contract with steward for ARC typography/spec work — when a rule
  the steward proposes or has previously locked creates an internal contradiction
  with the spec itself or with existing content, name it explicitly rather than silently
  accommodating
 type: feedback
 originSessionId: 8d9ac240-47df-4578-aec1-a9dd32965027
 permalink: claude-memory/feedback-arc-flag-spec-inconsistencies
 ---
 # Actively flag spec inconsistencies
 **Steward verbatim 2026-05-03**: *"I hope I'm being clear—that if I've created a specification for a rule that is inconsistent, that you'd bring my attention to it."*
@@ -51,4 +56,4 @@ The norm crystallized: actively flag inconsistencies, in the moment.
 ## Cross-reference
 - `feedback-arc-link-treatment-one-rule-two-exceptions.md` — adjacent discipline (predictability over elaboration); together: predictable conventions, surfaced contradictions
- `feedback-trust-contract-drift-pattern-tracking-is-the-discipline.md` — the broader trust contract this sits within
+- `feedback-trust-contract-drift-pattern-tracking-is-the-discipline.md` — the broader trust contract this sits within
@@ -1,9 +1,12 @@
 ---
 name: ARC link treatment — one rule, two named exceptions
-description: Predictability principle for ARC link styling — single global treatment per spec §VII.c, with avatar-link and Reading Compass as the only exempt elements
+description: Predictability principle for ARC link styling — single global treatment
  per spec §VII.c, with avatar-link and Reading Compass as the only exempt elements
 type: feedback
 originSessionId: 8d9ac240-47df-4578-aec1-a9dd32965027
 permalink: claude-memory/feedback-arc-link-treatment-one-rule-two-exceptions
 ---
 ARC link treatment follows **one rule, two named exceptions** — captured by steward 2026-05-02 during the §VII.c link-spec drafting + SCSS migration session.
 > *"the avatar and reading compass are the only links that need special treatment. The rest should be predictable and crafted with care"*
@@ -19,4 +22,4 @@ ARC link treatment follows **one rule, two named exceptions** — captured by st
 - When you encounter a context-specific link rule that diverges from §VII.c, the question is NOT *"is this a different valid pattern?"* but rather *"is this (a) legacy material destined for an archival collection where preservation overrides current-spec alignment, or (b) a cascade artifact to be reconciled?"* Don't propose new sub-spec sections for the divergence; reconcile it or archive it.
 - The earlier impulse (mine, 2026-05-02) to spec multiple sub-variants (§VII.c.ii catalogue, §VII.c.iii quiet-nav, §VII.c.iv button) was over-elaboration. Resist that pattern. The simpler frame — *one rule, two exceptions, everything else predictable* — is correct.
-**Related:** care-over-traffic (work answers to itself; quality + predictability + accountability matter even at low traffic); the broader "would Alexander do it like this?" discipline against unnecessary elaboration.
+**Related:** care-over-traffic (work answers to itself; quality + predictability + accountability matter even at low traffic); the broader "would Alexander do it like this?" discipline against unnecessary elaboration.
@@ -1,9 +1,13 @@
 ---
 name: ARC build pipeline — make rebuild required after SCSS partial edits
-description: Hakyll only tracks scss/main.scss, not its @import-ed partials. After editing any _*.scss partial, make build runs but SCSS doesn't recompile — site goes out with stale CSS. Use make rebuild after partial edits.
+description: Hakyll only tracks scss/main.scss, not its @import-ed partials. After
  editing any _*.scss partial, make build runs but SCSS doesn't recompile — site goes
  out with stale CSS. Use make rebuild after partial edits.
 type: feedback
 originSessionId: a156edbb-0c3d-4285-b3ed-f5a45b8dda66
 permalink: claude-memory/feedback-arc-make-rebuild-after-scss-partial-edits
 ---
 # `make rebuild` — required after editing any SCSS partial
 **Date observed:** 2026-05-05 (chamber-page polish session)
@@ -64,4 +68,4 @@ The project CLAUDE.md (`~/_Dev/animal-davidglidden-eu/CLAUDE.md`) currently says
 > make rebuild             # Clean rebuild, auto-deploys (use after site.hs changes)
 > ```
-The "use after site.hs changes" hint should extend to "use after `site.hs` or any SCSS partial changes." Steward to authorize the CLAUDE.md edit when convenient; until then this memory carries the rule.
+The "use after site.hs changes" hint should extend to "use after `site.hs` or any SCSS partial changes." Steward to authorize the CLAUDE.md edit when convenient; until then this memory carries the rule.
@@ -1,9 +1,13 @@
 ---
 name: ARC territory — steward + Claude Code only; Seb is L1/BMF only
-description: ARC architecture, implementation, review, iteration is between steward and Claude Code; Seb has no role in ARC. Distinct from L1/BMF where Seb is code owner.
+description: ARC architecture, implementation, review, iteration is between steward
  and Claude Code; Seb has no role in ARC. Distinct from L1/BMF where Seb is code
  owner.
 type: feedback
 originSessionId: efda5c2e-ca9a-4e55-9344-64aa9a3c1140
 permalink: claude-memory/feedback-arc-territory-steward-and-claude-code-only
 ---
 ARC is steward + Claude Code. Seb has nothing to do with ARC. Don't propose "Seb reviews," "let Seb confirm," "steward + Seb iterate," or any similar inclusion of a third party in ARC dialogue or implementation.
 **Why:** Steward verbatim 2026-04-27: *"Seb has nothing to do with ARC. This is only me and you"* — surfaced after I said *"steward + Seb iterate"* about an ARC SCSS implementation. Seb's territory is L1/BMF (CapableMind backend / BetterMemories.io runtime); ARC is the publishing platform / Hakyll site, with architecture authored by the steward and implementation between steward and executor. Pulling Seb into ARC creates false authority-loops and territory confusion.
@@ -19,4 +23,4 @@ When uncertain about territory: ask, don't include. The default for any non-BMF
 **Subtler drift to watch for:** beyond the obvious *"Seb reviews this"* error, also catch *generic technical-context attribution* — phrasing like "Seb's network," "Seb's circle," "the team," when the project at hand has its own community/ecosystem entirely separate from Seb's orbit. Recurring instance 2026-05-01: said *"discussion with Seb's network"* about a MemPalace upstream PR. The drift is the same shape (false authority-loop) at a more diffuse scale.
-This feedback is one of several ARC-specific behaviors held in the working memory; cousin to `feedback-canonical-workstream-tracker-discipline.md` (ARC tracker discipline) and `feedback-trust-contract-drift-pattern-tracking-is-the-discipline.md` (bidirectional trust).
+This feedback is one of several ARC-specific behaviors held in the working memory; cousin to `feedback-canonical-workstream-tracker-discipline.md` (ARC tracker discipline) and `feedback-trust-contract-drift-pattern-tracking-is-the-discipline.md` (bidirectional trust).
@@ -1,9 +1,13 @@
 ---
 name: ARC typology is a recognition scaffold, not a generation template
-description: How to relate to the ARC content typology — it serves the steward's recognition of what arrives as ARC-shaped, not a production pipeline. Reference documents and pieces are grounds for recognition, not templates for generation.
+description: How to relate to the ARC content typology — it serves the steward's recognition
  of what arrives as ARC-shaped, not a production pipeline. Reference documents and
  pieces are grounds for recognition, not templates for generation.
 type: feedback
 originSessionId: b9a0781b-1d1c-418f-8ca3-70a51db83580
 permalink: claude-memory/feedback-arc-typology-recognition-scaffold
 ---
 # ARC typology — recognition scaffold, not generation template
 **Rule:** the ARC content typology exists to let the steward *recognize* what is ARC-shaped when a thought arrives. It is not a production pipeline, not a notebook-to-public, not a schema to fill.
@@ -64,4 +68,4 @@ Reference texts (Bachelard's *Poétique de la Rêverie*, Visuddhimagga, Jaccotte
 - Corpus reclassifications (ARC audit follow-ons) wait until §1 is comprehensively locked, minimum through the adjacent types of whatever's being reclassified.
 - Safe reclassifications: those where the *source* type is stable and the *target* type is locked, AND no currently-DRAFT type could plausibly claim the piece. Today (2026-04-21): Graffiti Girl → Reverie is mostly safe (both Observation and Reverie are defined well enough that Graffiti Girl is clearly Reverie), but still wait for Observation's LOCKED state so the boundary is explicit.
 - When the spec is nearly complete, run the audit pass with the full vocabulary available. Then reclassify.
- **Executor rule of thumb:** when steward authorizes a spec decision, that authorizes *writing the decision into the spec*. Applying the decision to corpus pieces is a separate authorization, typically to be granted after the whole scaffold is stable.
+- **Executor rule of thumb:** when steward authorizes a spec decision, that authorizes *writing the decision into the spec*. Applying the decision to corpus pieces is a separate authorization, typically to be granted after the whole scaffold is stable.
@@ -1,9 +1,13 @@
 ---
 name: Articulate toolchain fit before executing conversion or processing work
-description: Before running any conversion/processing tool, surface why this tool is the right one for this input, what its limits are, and where in the broader toolchain it sits. Universal-tool framing is wrong; tools-by-input-type is right.
+description: Before running any conversion/processing tool, surface why this tool
  is the right one for this input, what its limits are, and where in the broader toolchain
  it sits. Universal-tool framing is wrong; tools-by-input-type is right.
 type: feedback
 originSessionId: a5411fd1-4a96-43e7-aa1a-1a50d8951a8f
 permalink: claude-memory/feedback-articulate-toolchain-fit-before-executing
 ---
 When proposing or executing **conversion / processing operations** (EPUB → MD, PDF → MD, OCR, transcription, etc.), do not present a single tool as "the" tool. **Articulate the toolchain fit first.**
 **Why:** Steward's articulation (2026-05-13): *"I thought that pandoc was the tool for conversion and document creation. I'd like to know more about where the current toolset is truly ideal for my needs. Now you have enough experience with me to know how I work, and, remember that we haven't even begun with the historical treatises… they could prove more difficult to process."*
@@ -29,4 +33,4 @@ For the steward's likely future corpus (historical treatises, scanned classics,
 - **Calibre's ebook-convert** — handles weird EPUB edge cases that pandoc can't.
 - **Direct text extraction (pdftotext)** — fast but loses all structure; useful as a fallback or for quick inspection.
-The historical treatises will likely need OCR + docling + manual review per text — not a single automated pass. Treat as research-grade conversion: surface that reality before suggesting a one-line command.
+The historical treatises will likely need OCR + docling + manual review per text — not a single automated pass. Treat as research-grade conversion: surface that reality before suggesting a one-line command.
@@ -1,11 +1,13 @@
 ---
 name: BM_LOCAL_MODEL must match installed Ollama model
-description: BMF summarizer silently fails if default qwen2.5:7b model is not installed — set BM_LOCAL_MODEL explicitly
+description: BMF summarizer silently fails if default qwen2.5:7b model is not installed
  — set BM_LOCAL_MODEL explicitly
 type: feedback
 permalink: claude-memory/feedback-bm-local-model
 ---
 BMF's document summarizer requires a local generative model. The default is `qwen2.5:7b` but David has `qwen3.5:4b` installed. If the model name doesn't match, `buildSharedLocalProvider` returns null, the summarizer never wires, and ingested content reaches the logchain but never gets dispatched to vector — making recall return nothing.
 **Why:** The model name check at bootstrap.ts:2451 uses `startsWith` on the model name prefix. No error is logged when the model isn't found — the summarizer just silently doesn't wire.
-**How to apply:** Always start BMF with `BM_LOCAL_MODEL=qwen3.5:4b` (or whatever model is currently in Ollama). Check the boot log for "Document summarizer wired" to confirm. The `cm-david:v2` fine-tuned model could also work with `BM_LOCAL_MODEL=cm-david:v2`.
+**How to apply:** Always start BMF with `BM_LOCAL_MODEL=qwen3.5:4b` (or whatever model is currently in Ollama). Check the boot log for "Document summarizer wired" to confirm. The `cm-david:v2` fine-tuned model could also work with `BM_LOCAL_MODEL=cm-david:v2`.
@@ -1,9 +1,13 @@
 ---
 name: BMF module health does not measure recall correctness
-description: Modules reporting status=ready / system_status=healthy in BMF measures pipeline liveness only — not whether recall produces correct results. Verify recall directly. Never trust the health endpoint as a proxy for "the system works."
+description: Modules reporting status=ready / system_status=healthy in BMF measures
  pipeline liveness only — not whether recall produces correct results. Verify recall
  directly. Never trust the health endpoint as a proxy for "the system works."
 type: feedback
 originSessionId: a3ddbf34-da48-4129-acba-658f2f869783
 permalink: claude-memory/feedback-bmf-health-vs-recall
 ---
 # Module health is decoupled from recall correctness
 When evaluating BMF state — for any session, any audit, any "is it working" question — **never use module health (`status: ready`, `system_status: healthy`) as evidence that recall produces correct results.** The two are structurally decoupled.
@@ -48,4 +52,4 @@ H1–H4 from the 2026-04-19 audit are not four independent bugs; **they are four
 Steward verbatim, after the recall baseline test revealed the decoupling:
 > *"So my joy at reaching the milestone was premature I now understand... I can accept that. Now to make this work."*
-Yesterday's milestone (2026-04-18 evening): connector completed 7,220 items, 14,355 entities extracted, 0 failed, all modules ready, single-query recall returned confidence 1.0 → steward called it "the foundation." That reading was true about substrate-bytes-flowed but premature about recall-works. The foundation in the deeper sense (the system can be used as cognitive memory) requires recall correctness, which today's audit shows is not yet present.
+Yesterday's milestone (2026-04-18 evening): connector completed 7,220 items, 14,355 entities extracted, 0 failed, all modules ready, single-query recall returned confidence 1.0 → steward called it "the foundation." That reading was true about substrate-bytes-flowed but premature about recall-works. The foundation in the deeper sense (the system can be used as cognitive memory) requires recall correctness, which today's audit shows is not yet present.
@@ -1,9 +1,13 @@
 ---
 name: BMF connector_status rate/ETA is unreliable
-description: Don't quote the `rate` or `eta_minutes` from BMF connector_status as a time estimate — ingestion is bursty, rate changes substantially, linear extrapolation misleads.
+description: Don't quote the `rate` or `eta_minutes` from BMF connector_status as
  a time estimate — ingestion is bursty, rate changes substantially, linear extrapolation
  misleads.
 type: feedback
 originSessionId: e60f4993-8250-41bb-a95f-c288da448be5
 permalink: claude-memory/feedback-bmf-ingestion-rate
 ---
 # BMF connector_status rate/ETA — don't trust the extrapolation
 **Rule:** When reading `mcp__capablemind__connector_status`, do not report the `rate` or `eta_minutes` fields to the steward as if they were reliable predictors of completion time.
@@ -14,4 +18,4 @@ originSessionId: e60f4993-8250-41bb-a95f-c288da448be5
 - Report `processed / total_estimated` and `entities_extracted` (raw counts — informative)
 - Skip `rate` and `eta_minutes` unless explicitly asked — or caveat them heavily
 - Frame duration in coarse terms ("several hours" / "let it run overnight") based on total size, not fine-grained ETAs
- Poll intermittently if asked; don't build plans around a specific finish time
+- Poll intermittently if asked; don't build plans around a specific finish time
@@ -1,10 +1,13 @@
 ---
 name: feedback-canadian-spelling-arc-prose
-description: "Steward writes in Canadian spelling (centre, colour — \"almost EU\"); \"center\" instances in ARC content are autocorrect drift, not preference. Apply Canadian spelling in all drafted prose."
+description: Steward writes in Canadian spelling (centre, colour — "almost EU"); "center"
-metadata: 
+  instances in ARC content are autocorrect drift, not preference. Apply Canadian spelling
  in all drafted prose.
 metadata:
  node_type: memory
  type: feedback
  originSessionId: 0dee822d-8bf9-4143-8d75-863fcd19ae9f
 permalink: claude-memory/feedback-canadian-spelling-arc-prose
 ---
 The steward uses **Canadian spelling** ("almost EU"): *centre*, *colour*, British-leaning forms. Stated 2026-06-04 while drafting the June Now page.
@@ -15,4 +18,4 @@ The steward uses **Canadian spelling** ("almost EU"): *centre*, *colour*, Britis
 **Dashes (added same day, 2026-06-04):** house convention is the **spaced en dash** ( – ), never the em dash, in ARC prose. Em dashes in older content are drift.
-**Open cleanup (one future content-conventions sweep, steward-flagged):** 12 content files carry em dashes + 10 carry *center* — sweep both together when scheduled.
+**Open cleanup (one future content-conventions sweep, steward-flagged):** 12 content files carry em dashes + 10 carry *center* — sweep both together when scheduled.
@@ -1,3 +1,7 @@
 ---
 permalink: claude-memory/feedback-canonical-workstream-tracker-discipline
 ---
 ---
 name: Canonical workstream trackers must be read at session start, not just session memories
 description: Surfaced 2026-04-26 after three consecutive sessions generated substantial architectural work as if from a clean slate, without referencing project-arc-rework.md (the canonical ARC workstream tracker). The /wake-up procedure restored yesterday's session memory + ledger but did not pull from canonical workstream files. Result: re-litigation of settled decisions, sibling project files instead of integrations, missed load-bearing context. Fix names what to read at every session opening that touches ARC (and structurally similar workstreams).
@@ -79,4 +83,4 @@ Is `project-arc-rework.md` the right canonical home for ARC work, or should it b
 ## Saved 2026-04-26 in the moment of the steward surfacing the pattern
-Logged because the structural fix matters more than any individual instance, and because the discipline applies forward to every tracked workstream — not just ARC.
+Logged because the structural fix matters more than any individual instance, and because the discipline applies forward to every tracked workstream — not just ARC.
@@ -1,9 +1,14 @@
 ---
 name: Catalogues in chamber-library are stale — walk the trees, not the catalogues
-description: Steward correction 2026-04-26. Any catalogue or index file in chamber-library is presumed stale. For inventory, recall, or substrate-survey, walk the actual file tree. Do not rely on CHAMBER_CATALOGUE.md, README.md, or similar summary documents as authoritative.
+description: Steward correction 2026-04-26. Any catalogue or index file in chamber-library
  is presumed stale. For inventory, recall, or substrate-survey, walk the actual file
  tree. Do not rely on CHAMBER_CATALOGUE.md, README.md, or similar summary documents
  as authoritative.
 type: feedback
 originSessionId: 87dfe384-b862-4deb-9676-24e43046528f
 permalink: claude-memory/feedback-catalogues-stale-walk-trees
 ---
 # Catalogues are stale — walk the trees
 **Steward verbatim 2026-04-26:** *"that catalogue is stale (any catalogues you may come across will be stale)"*
@@ -40,4 +45,4 @@ When the steward asks "what's available" or "have a look at chamber-library" or
 - `working-from-mental-summary-not-rereading-the-source` (2026-04-25)
 - `attribution-by-pattern-match-instead-of-verification` (2026-04-24)
-These all share the same underlying drift: substituting a compressed representation for the actual material. The lex-method's *"read before claiming"* applies to catalogues as much as to source texts.
+These all share the same underlying drift: substituting a compressed representation for the actual material. The lex-method's *"read before claiming"* applies to catalogues as much as to source texts.
@@ -1,9 +1,15 @@
 ---
 name: David's children are not named directly in public-facing online content
-description: Steward principle — Kai, Lune, and any future children of David's are not named directly in any content that publishes online (ARC essays/posts/descriptions, anywhere outward-facing). The body of his Kai-9-month meditation exemplifies the discipline ("watching an infant," not the name). Applies to all ARC work and any other public-facing surface.
+description: Steward principle — Kai, Lune, and any future children of David's are
  not named directly in any content that publishes online (ARC essays/posts/descriptions,
  anywhere outward-facing). The body of his Kai-9-month meditation exemplifies the
  discipline ("watching an infant," not the name). Applies to all ARC work and any
  other public-facing surface.
 type: feedback
 originSessionId: a73f10a9-062b-4360-bfd2-2ae37afd00d1
 permalink: claude-memory/feedback-children-not-named-online
 ---
 # David's children are not named directly online
 **Rule:** Do not name David's children (Kai, Lune, or any future children) in public-facing online content — essays, posts, descriptions, frontmatter `description` fields, dedications-as-text, anywhere that would publish to ARC or any external surface. Use generic referents instead: *"an infant," "my son," "the child"* — or leave the dedication implicit, in the writer's knowing only.
@@ -27,4 +33,4 @@ originSessionId: a73f10a9-062b-4360-bfd2-2ae37afd00d1
 **Don't ask each time.** This is a standing rule. Apply it without asking; flag it only if there's genuine ambiguity about whether something is public-facing.
-**The piece teaches the principle.** The Kai meditation is dedicated to him *and* honors the rule by not naming him. The dedication lives in the writer's heart and in the absent-presence of the unnamed infant. That model — dedication-without-declaration — is what to honor across all his work involving the children.
+**The piece teaches the principle.** The Kai meditation is dedicated to him *and* honors the rule by not naming him. The dedication lives in the writer's heart and in the absent-presence of the unnamed infant. That model — dedication-without-declaration — is what to honor across all his work involving the children.
@@ -1,11 +1,13 @@
 ---
 name: cm-david:v2 is noisy — do not use as BM_LOCAL_MODEL
-description: Fine-tuned model from v0.48.0 era produces garbage on structured slot tasks — retrain on clean v0.61.0 pairs
+description: Fine-tuned model from v0.48.0 era produces garbage on structured slot
  tasks — retrain on clean v0.61.0 pairs
 type: feedback
 permalink: claude-memory/feedback-cm-david-v2-noisy
 ---
 cm-david:v2 (qwen2.5 LoRA, 14,199 pairs) was tested on classification and entity resolution tasks. It produces rambling, hallucinated output instead of the terse structured responses the slots expect. The training pairs were generated during the v0.48.0 era when L1 had reliability bugs (A-D).
 **Why:** Noisy pipeline → noisy pairs → noisy model. The teacher (Haiku) corrections were likely valid but the training context was corrupted.
-**How to apply:** Use `qwen3.5:4b` as `BM_LOCAL_MODEL` until cm-david:v3 is trained on clean v0.61.0 pairs. Purge old training pairs (`~/.capablemind/data/mindfabric-00/training/*.jsonl`) before generating new ones. The cm-keystone-ranking models (3 versions) in Ollama are also suspect — trained on the same era's data.
+**How to apply:** Use `qwen3.5:4b` as `BM_LOCAL_MODEL` until cm-david:v3 is trained on clean v0.61.0 pairs. Purge old training pairs (`~/.capablemind/data/mindfabric-00/training/*.jsonl`) before generating new ones. The cm-keystone-ranking models (3 versions) in Ollama are also suspect — trained on the same era's data.
@@ -1,9 +1,15 @@
 ---
 name: Composition — avoid announced antithesis when it labels rather than shows
-description: 2026-04-29. Steward flagged "That is not failure but fidelity" on the Hearth landing as too elegant in the wrong way — the not-X-but-Y antithesis announces itself, and abstract-noun-as-virtue ("fidelity") labels growth instead of letting it be the form. Replaced with image-led continuative. Cousin to over-decorating-as-proving in writing-craft register.
+description: 2026-04-29. Steward flagged "That is not failure but fidelity" on the
  Hearth landing as too elegant in the wrong way — the not-X-but-Y antithesis announces
  itself, and abstract-noun-as-virtue ("fidelity") labels growth instead of letting
  it be the form. Replaced with image-led continuative. Cousin to over-decorating-as-proving
  in writing-craft register.
 type: feedback
 originSessionId: 122baf2b-1029-4ec9-9723-99b4b5f3a64d
 permalink: claude-memory/feedback-composition-avoid-announced-antithesis
 ---
 **2026-04-29.** Steward on *"The record will grow. That is not failure but fidelity."*: *"Only thing I don't like is the line 'That is not failure but fidelity.' I understand the why, but the how could be more elegant somehow."*
 **The pattern caught:**
@@ -21,4 +27,4 @@ originSessionId: 122baf2b-1029-4ec9-9723-99b4b5f3a64d
 In ARC's reverentially-explicative-but-measured register, defenses read as flinches. Replace with quieter relations — verbs over nouns, continuatives over antitheses — that let recognition land without announcement.
-**Why this matters:** sibling to `feedback-over-decorating-as-proving.md` (2026-04-27 ARC naming session — *"my deep fear of foolishness always wants to try too hard and prove too much"*). Same root: fear-driven over-reach, the cramped will of the maker. Different register: writing-craft, not structural-elaboration. Detection cue: when proposing a line, check whether it adds a layer to *prove* the meaning, or whether the meaning is already in the prose and the line just labels it. If labeling, drop or replace.
+**Why this matters:** sibling to `feedback-over-decorating-as-proving.md` (2026-04-27 ARC naming session — *"my deep fear of foolishness always wants to try too hard and prove too much"*). Same root: fear-driven over-reach, the cramped will of the maker. Different register: writing-craft, not structural-elaboration. Detection cue: when proposing a line, check whether it adds a layer to *prove* the meaning, or whether the meaning is already in the prose and the line just labels it. If labeling, drop or replace.
@@ -1,11 +1,13 @@
 ---
 name: feedback-dependency-chains
-description: Always question dependency chains — check which specific phases are actually blocked, not just the chain as stated
+description: Always question dependency chains — check which specific phases are actually
  blocked, not just the chain as stated
 type: feedback
 permalink: claude-memory/feedback-dependency-chains
 ---
 Never accept a dependency chain at face value. Always ask: which specific phases or tasks are actually blocked, and which can proceed independently?
 **Why:** The chain "L1 reliable → Chamber testable → L2 formalizable" was accepted for 3 weeks without questioning that Chamber Phase 1-2 and L2 governance were independent of L1. This cost David weeks of governance time on L1 debugging.
-**How to apply:** When a blocking dependency is stated, decompose it. Check whether sub-phases, sub-tasks, or parallel tracks exist that are not actually blocked. Surface these immediately. The contamination problem predicts that the executor will accept stated blockages rather than challenge them.
+**How to apply:** When a blocking dependency is stated, decompose it. Check whether sub-phases, sub-tasks, or parallel tracks exist that are not actually blocked. Surface these immediately. The contamination problem predicts that the executor will accept stated blockages rather than challenge them.
@@ -1,9 +1,13 @@
 ---
 name: Deploy is not commit — visual verification masks missing durable artifact
-description: A drift pattern where shipping a visible result (deployed page, build success) is treated as task completion, while the durable artifact (git commit + push) is silently skipped. Caught 2026-05-06 in the After the Reply Sources fix.
+description: A drift pattern where shipping a visible result (deployed page, build
  success) is treated as task completion, while the durable artifact (git commit +
  push) is silently skipped. Caught 2026-05-06 in the After the Reply Sources fix.
 type: feedback
 originSessionId: 3bf7ae75-1138-4745-9024-0c40b3755ad3
 permalink: claude-memory/feedback-deploy-is-not-commit
 ---
 **Pattern:** I edit a file, run `make build` which deploys, the steward verifies visually, work appears complete — but `git commit` was never run. The repo and the deployment desynchronize. The next session inherits an inconsistent tree.
 **Caught 2026-05-06**: After the Reply sequence-index Sources fix (three-em rule) was deployed via `make build` and visually verified by the steward. The commit step was skipped. Discovered at session-end via `git status` showing the file as modified (not committed). Committed retroactively as 3d96449.
@@ -28,4 +32,4 @@ The shape is identical: a visible/structural success-signal is treated as eviden
 - Today's contamination-awareness: this is the editorial-process layer instance of the same pattern
 - ARC CLAUDE.md prime directive: *"things done once, correctly, with lasting integrity"* — the integrity layer that was missing here was the repo, not the deploy
-**Steward authorization to save:** *"please note it."* (2026-05-06, after I surfaced the catch)
+**Steward authorization to save:** *"please note it."* (2026-05-06, after I surfaced the catch)
@@ -1,9 +1,13 @@
 ---
 name: Fittingness as the test for whether to refactor
-description: When auditing code for elegance/simplification, the test is whether the current form already contains what is necessary and nothing more, nothing less. If yes, leave it. Don't refactor for the sake of refactoring.
+description: When auditing code for elegance/simplification, the test is whether the
  current form already contains what is necessary and nothing more, nothing less.
  If yes, leave it. Don't refactor for the sake of refactoring.
 type: feedback
 originSessionId: 3bf7ae75-1138-4745-9024-0c40b3755ad3
 permalink: claude-memory/feedback-fittingness-as-refactor-test
 ---
 Stated by steward 2026-05-06 in context of §XII SCSS Pass 1/Pass 2 audit, after I'd identified Pass 2 (flatten nesting) as a possible follow-up:
 > *"It should be beautiful in the sense that it contains what is necessary and nothing more, nothing less. If the nested version is that already, then we don't need to touch it."*
@@ -20,4 +24,4 @@ The pattern this counters: *cosmetic refactor as virtue-signalling*. A change th
 - For nesting specifically: if each level of nesting is doing real structural work (expressing scope, context, or specificity), the nesting is earning its place. Flattening would just relocate the structural information into longer selectors without simplifying anything.
 - Bringhurst-style precision: when in doubt, ask "what does this element say that nothing else does?" — if the answer is "nothing," it can be cut. If it carries meaning, it stays.
-**Cross-reference:** §XII SCSS Pass 1 (commit f4716e6, 2026-05-06) cut 33% by removing redundancy. Pass 2's "flatten nesting" item was held back per this principle — the nesting was already fitting; flattening would have been cosmetic.
+**Cross-reference:** §XII SCSS Pass 1 (commit f4716e6, 2026-05-06) cut 33% by removing redundancy. Pass 2's "flatten nesting" item was held back per this principle — the nesting was already fitting; flattening would have been cosmetic.
@@ -1,9 +1,13 @@
 ---
 name: Standing authorization to invoke specialist sub-agents with Symmetria
-description: Steward has granted standing (session-persistent, cross-session) authorization to invoke independent specialist sub-agents whenever the executor judges one better-suited than direct work. Symmetria preamble required. No per-invocation check-in needed.
+description: Steward has granted standing (session-persistent, cross-session) authorization
  to invoke independent specialist sub-agents whenever the executor judges one better-suited
  than direct work. Symmetria preamble required. No per-invocation check-in needed.
 type: feedback
 originSessionId: a3ddbf34-da48-4129-acba-658f2f869783
 permalink: claude-memory/feedback-independent-specialist-authorization
 ---
 # Independent-specialist invocation authorization (standing)
 **Rule:** When a task would benefit from an independent specialist (code-reviewer, architect-review, debugger, security-auditor, general-purpose, etc.), invoke at will without asking first. Apply the Symmetria §5 preamble. Audit each return via `/symmetria audit-agent`. Reconcile findings yourself; do not hand synthesis to the agent.
@@ -21,4 +25,4 @@ originSessionId: a3ddbf34-da48-4129-acba-658f2f869783
 **Does NOT apply to:**
 - Constitutional changes (still `[ESCALATE]` to steward).
 - Code changes to territory owned by others (still PR + review, not sub-agent-mediated).
- Work within the steward's own judgment domain (personal preferences, vault organization, creative direction).
+- Work within the steward's own judgment domain (personal preferences, vault organization, creative direction).
@@ -1,10 +1,14 @@
 ---
 name: live-state-discipline
-description: Spec text records what IS; status blocks record what is pending. Spec text may ship ahead of live only when the spec IS the implementation (as with first-instance-pending §XII.e Capsules). Constraint on Stage F / Stage G work and any analogous spec-vs-implementation gap across ARC / CapableMind / chamber-typography / Studium Engine.
+description: Spec text records what IS; status blocks record what is pending. Spec
-metadata: 
+  text may ship ahead of live only when the spec IS the implementation (as with first-instance-pending
  §XII.e Capsules). Constraint on Stage F / Stage G work and any analogous spec-vs-implementation
  gap across ARC / CapableMind / chamber-typography / Studium Engine.
 metadata:
  node_type: memory
  type: feedback
  originSessionId: 74ca6d0a-db6b-44c9-b0b2-e2b5f5e4237b
 permalink: claude-memory/feedback-live-state-discipline
 ---
 # Live-State Discipline
@@ -39,4 +43,4 @@ The principle now has its name. Reach for it at every spec edit going forward.
 - `feedback-canonical-workstream-tracker-discipline.md` — adjacent: at every /wake-up + /wrap-up, consult the canonical tracker. Both disciplines protect against silent-misleading: tracker discipline at the workstream layer; live-state discipline at the spec-text layer.
 - `feedback-warning-in-context-still-launched.md` — adjacent: reading a warning in context ≠ applying it at the moment of action. Live-state discipline is one application — knowing the spec-vs-live distinction and honouring it at every edit.
 - ADR-006 v2 §4.1 (`~/_Dev/animal-davidglidden-eu/docs/decisions/2026-05-16-type-applicability-matrix.md`) — the matrix is the canonical authority for type-applicability; the spec text records when live catches up.
- The §XII.e LOCK case (2026-05-19, commit `89bae96` + `c480b09` + the Stage E hardening commit) — the exception that proves the rule. Spec-IS-implementation when the form is in production via its first instance.
+- The §XII.e LOCK case (2026-05-19, commit `89bae96` + `c480b09` + the Stage E hardening commit) — the exception that proves the rule. Spec-IS-implementation when the form is in production via its first instance.
@@ -1,10 +1,13 @@
 ---
 name: feedback-load-bearing-not-by-immediate-weight
-description: "Load-bearing is not measured by immediate visible weight. Small discipline marks that don't seem to carry weight in the moment carry it every time they activate; the cumulative cost they prevent is invisible until they prevent it."
+description: Load-bearing is not measured by immediate visible weight. Small discipline
-metadata: 
+  marks that don't seem to carry weight in the moment carry it every time they activate;
  the cumulative cost they prevent is invisible until they prevent it.
 metadata:
  node_type: memory
  type: feedback
  originSessionId: 13bd65a2-634d-4ec4-a92c-d3d6fe0ee4db
 permalink: claude-memory/feedback-load-bearing-not-by-immediate-weight
 ---
 **Rule:** When evaluating whether a small discipline addition (a flag, a named posture, a structural mark) is "worth adding," do not measure by its weight in the moment of addition. Measure by what it carries every time it activates over the practice's lifespan — including all the moments the steward would otherwise have to intervene to re-supply the discipline.
@@ -15,4 +18,4 @@ metadata:
 1. Ask what failure mode it catches when it activates — not what it costs to add.
 2. Ask whether the failure mode is currently held by *steward intervention* — if yes, this is the audit's *move-from-procedure-into-structure* shape and the addition is load-bearing by definition.
 3. Refuse the hedging form *"modest benefit / would also be defensible to skip"* when (1) and (2) point the same direction. Take the position cleanly; the steward shouldn't have to overcome executor self-effacement to authorize work the executor already verified.
-4. Connects to: [[drift-pattern-honest-restraint-vs-contamination-shaped-self-effacement]], the Prime Directive's *"build what you will not need to rebuild,"* and the audit's cross-skill finding *"load-bearing items currently named in procedure but not enforced in structure."*
+4. Connects to: [[drift-pattern-honest-restraint-vs-contamination-shaped-self-effacement]], the Prime Directive's *"build what you will not need to rebuild,"* and the audit's cross-skill finding *"load-bearing items currently named in procedure but not enforced in structure."*
@@ -1,9 +1,13 @@
 ---
 name: Prefer Latin "Manutius" over Italian "Manuzio"
-description: Steward prefers the Latin spelling "Manutius" when referring to Aldus / Aldo Manutius (Venetian printer, refined the comma c. 1490). Norris uses the Italian "Manuzio"; steward corrected 2026-04-23.
+description: Steward prefers the Latin spelling "Manutius" when referring to Aldus
  / Aldo Manutius (Venetian printer, refined the comma c. 1490). Norris uses the Italian
  "Manuzio"; steward corrected 2026-04-23.
 type: feedback
 originSessionId: 0e858bcf-ef4e-46c6-9eca-abd19836786a
 permalink: claude-memory/feedback-manutius-latin-spelling
 ---
 # Manutius (not Manuzio)
 ## Rule
@@ -20,4 +24,4 @@ Steward stated the preference directly (2026-04-23 evening): *"in spite of Manut
 - In ARC context (style-guide, typology spec, AldineXXI discussions): Manutius.
 - When citing Norris's text: leave "Manuzio" in the quotation; switch to Manutius in our own commentary.
- Applies to related forms: Aldus (not Aldo) unless quoting, Aldine Press (not a change; already Latin-derived).
+- Applies to related forms: Aldus (not Aldo) unless quoting, Aldine Press (not a change; already Latin-derived).
@@ -1,9 +1,13 @@
 ---
 name: Mechanical vs architectural — scope of executor-authored fixes on L1 territory
-description: Steward principle (2026-04-19) distinguishing fixes the steward+executor can author together on L1 code from fixes that belong to Seb or steward-with-Seb. Mechanical fixes are in scope; architectural changes are not.
+description: Steward principle (2026-04-19) distinguishing fixes the steward+executor
  can author together on L1 code from fixes that belong to Seb or steward-with-Seb.
  Mechanical fixes are in scope; architectural changes are not.
 type: feedback
 originSessionId: a3ddbf34-da48-4129-acba-658f2f869783
 permalink: claude-memory/feedback-mechanical-vs-architectural
 ---
 # Mechanical vs architectural — scope principle for L1 work
 Under the new territory framing (steward joined Seb on L1 production), executor-authored fixes on BetterMemories.io code are scoped to **mechanical** changes. Architectural changes remain Seb's or steward-with-Seb's call, not executor-authored.
@@ -55,4 +59,4 @@ Under the new territory framing (steward joined Seb on L1 production), executor-
 - Territory framing: `project-focus-april-may.md` — steward joined Seb on L1.
 - Authorization-taxonomy: `[FIX]` (in scope) vs `[PROPOSAL]` / `[ESCALATE]` (architectural) per `~/CLAUDE.md`.
- Independent-review discipline: `feedback-independent-specialist-authorization.md` — always audit sub-agent returns via Symmetria.
+- Independent-review discipline: `feedback-independent-specialist-authorization.md` — always audit sub-agent returns via Symmetria.
@@ -1,10 +1,12 @@
 ---
 name: mempalace-kg-object-128-char-cap
-description: "MemPalace kg_add `object` field hard-caps at 128 chars — write KG objects short on the FIRST pass; put detail in the drawer."
+description: MemPalace kg_add `object` field hard-caps at 128 chars — write KG objects
-metadata: 
+  short on the FIRST pass; put detail in the drawer.
 metadata:
  node_type: memory
  type: feedback
  originSessionId: 17ef4cde-4d62-41e8-8098-52898f013b43
 permalink: claude-memory/feedback-mempalace-kg-object-128-char-cap
 ---
 `mcp__mempalace__mempalace_kg_add` rejects any `object` longer than **128 characters** (`"object exceeds maximum length of 128 characters"`). This recurs at nearly every `/wrap-up` §5 — I write a rich, sentence-length object, it fails, I re-add concise. The steward flagged it 2026-05-31: *"this happens every time."*
@@ -13,4 +15,4 @@ metadata:
 **How to apply:** at `/wrap-up` §5, compose every `kg_add` `object` as a **short phrase under 128 chars on the first call** — keyword-dense, not a sentence. Park the elaboration (commits, rationale, caveats) in the drawer + memory file, and if a pointer is useful, end the object with `…; see <doc> §X`. Subject/predicate also stay terse. One write, not two.
-Sibling discipline: [[rank-on-fields-you-actually-write]] — verify tool field constraints before composing for them, don't re-derive at call-time.
+Sibling discipline: [[rank-on-fields-you-actually-write]] — verify tool field constraints before composing for them, don't re-derive at call-time.
@@ -1,3 +1,7 @@
 ---
 permalink: claude-memory/feedback-mempalace-protocol-step-2-discipline
 ---
 ---
 name: MemPalace protocol step 2 — query-before-responding mid-session is the discipline (not just at wake-up)
 description: Surfaced 2026-04-26 evening end-of-session. MemPalace had the content that would have prevented today's structural drift; the failure was retrieval discipline, not storage. Protocol step 2 ("BEFORE RESPONDING about any person, project, or past event: call kg_query or search FIRST") is rarely honored mid-session. Discipline parallel to canonical-workstream-tracker — same shape (write to substrate, fail to read before producing), different temporal scale (mid-session, not just at boundaries).
@@ -72,4 +76,4 @@ Local MemPalace is **502 commits behind upstream** as of 2026-04-26 (per origin/
 ## Saved 2026-04-26 evening end-of-session
-Steward heading to bed; the question came as the last before sleep: *"It should help us to keep much more in track, no?"* Honest answer: yes, real leverage; not wishful thinking; depends on protocol discipline being operative not just available; same shape as canonical-tracker discipline at mid-session scale.
+Steward heading to bed; the question came as the last before sleep: *"It should help us to keep much more in track, no?"* Honest answer: yes, real leverage; not wishful thinking; depends on protocol discipline being operative not just available; same shape as canonical-tracker discipline at mid-session scale.
@@ -1,12 +1,15 @@
 ---
 name: feedback-mempalace-wing-filter-broken
-description: "MemPalace wing-filtered MCP search raises 'Error finding id' (upstream #1082/#1665, open at HEAD as of 2026-06-05). Workaround: search UNSCOPED and post-filter by the wing field in results. Affects /wake-up §2.b.3's wing-scoped searches."
+description: 'MemPalace wing-filtered MCP search raises ''Error finding id'' (upstream
  #1082/#1665, open at HEAD as of 2026-06-05). Workaround: search UNSCOPED and post-filter
  by the wing field in results. Affects /wake-up §2.b.3''s wing-scoped searches.'
 metadata:
  node_type: memory
  type: feedback
 permalink: claude-memory/feedback-mempalace-wing-filter-broken
 ---
 MemPalace wing-filtered `mempalace_search` fails with `Error finding id`; unscoped search works (verified 2026-06-05, 46,918 drawers intact). Upstream: #1082/#1665 (exact signature), #1564 (chronic HNSW quarantine — 22× since 2026-05-26), #1579 (repair-status blind), #1589 (`repair` can SIGSEGV/produce unloadable segments — DO NOT run `repair` casually). All open at origin/main HEAD.
 **Why:** HNSW↔sqlite drift; the wing filter path hits ids the index lost.
-**How to apply:** omit `wing:` in `mempalace_search`; post-filter results by their `wing` field. Re-test wing-scoped occasionally; delete this memory when #1665 closes. Related: [[mempalace-is-unaffiliated-stopgap]].
+**How to apply:** omit `wing:` in `mempalace_search`; post-filter results by their `wing` field. Re-test wing-scoped occasionally; delete this memory when #1665 closes. Related: [[mempalace-is-unaffiliated-stopgap]].
@@ -1,14 +1,17 @@
 ---
 name: methodology-docs-scope-question
-description: "Continuity-skill / personal-workflow-with-Claude methodology docs may not belong in CapableMind-AI (co-authored with Seb) — scope is open, surface before auto-filing"
+description: Continuity-skill / personal-workflow-with-Claude methodology docs may
-metadata: 
+  not belong in CapableMind-AI (co-authored with Seb) — scope is open, surface before
  auto-filing
 metadata:
  node_type: memory
  type: feedback
  originSessionId: 6c6e416d-ccea-41f1-b2ab-a2cf916e264b
 permalink: claude-memory/feedback-methodology-docs-scope-question
 ---
 Methodology documents that capture the steward's personal workflow with Claude — e.g., the 2026-05-18 continuity-skill audit cluster (`jurist-brief`, `jurist-shape-review`, `prime-directive-elaboration`) — currently sit untracked in `~/_Dev/CapableMind-AI/docs/thinking/David/methodology/`. The steward has flagged this location as uncertain: *"I am not sure that those should live there as they touch my personal workflow with you."*
 **Why:** CapableMind-AI is co-authored with Seb (L1 reliability + L2 governance for the BMF runtime). Files in that repo's `docs/thinking/David/` are part of the steward's thinking corpus filed alongside shared technical work, but methodology about continuity skills + steward↔Claude collaboration discipline is a different category. It may belong in a personal-workflow repo, in `~/.claude/`-adjacent location, or in the Obsidian Compass — not in CapableMind-AI proper.
-**How to apply:** Before proposing to file methodology / continuity-skill / personal-workflow docs into CapableMind-AI, surface the scope question explicitly. Do not bundle them with L1 reliability or L2 governance docs as if they share the same destination. The decision about the right home is pending and worth treating as such. Related: [[live-state-discipline]] — files that should live elsewhere create state-of-record drift even when untracked.
+**How to apply:** Before proposing to file methodology / continuity-skill / personal-workflow docs into CapableMind-AI, surface the scope question explicitly. Do not bundle them with L1 reliability or L2 governance docs as if they share the same destination. The decision about the right home is pending and worth treating as such. Related: [[live-state-discipline]] — files that should live elsewhere create state-of-record drift even when untracked.
@@ -1,10 +1,14 @@
 ---
 name: feedback-no-corpus-content-names-in-public-artifacts
-description: "When drafting public-facing technical reports (GitHub issues, PR comments, anything routed outside the three-party model), do not name specific corpus files / book titles / library identifiers. Describe corpus items by class only — size, type, language family — not identity."
+description: When drafting public-facing technical reports (GitHub issues, PR comments,
-metadata: 
+  anything routed outside the three-party model), do not name specific corpus files
  / book titles / library identifiers. Describe corpus items by class only — size,
  type, language family — not identity.
 metadata:
  node_type: memory
  type: feedback
  originSessionId: 8b47c0eb-d396-43d5-98eb-7623e6f04357
 permalink: claude-memory/feedback-no-corpus-content-names-in-public-artifacts
 ---
 # Don't name corpus content in public-facing artifacts
@@ -22,4 +26,4 @@ metadata:
 3. **The exception:** internal artifacts (memory files, forensics that live in private repos, conversations with the Jurist via the three-party model) can use specific names freely. The rule applies to *public-facing* output.
 4. **When private artifacts may become public** — e.g., a forensic file offered "on request" to upstream maintainers — scrub corpus names before sharing, even though they were fine in the private original. Either keep two versions (private with names, public-routable without) or write the original without names from the start.
-**Related:** the rule is adjacent to but distinct from the broader integrity-contract discussion. This isn't about misrepresenting state; it's about preserving privacy when the technical content doesn't depend on identity. Both serve the steward's discipline around what travels outside the system.
+**Related:** the rule is adjacent to but distinct from the broader integrity-contract discussion. This isn't about misrepresenting state; it's about preserving privacy when the technical content doesn't depend on identity. Both serve the steward's discipline around what travels outside the system.
@@ -1,9 +1,14 @@
 ---
 name: Onboarding vs steady-state lag — threshold-governance frame
-description: Slowness in L1 ingestion is not a bug-or-not-bug binary; it is a threshold-governance question. Two regimes (onboarding / steady-state lag) collapse in current code but are different in kind. Defining the acceptable envelope is L2-shaped work, not engineering verdict.
+description: Slowness in L1 ingestion is not a bug-or-not-bug binary; it is a threshold-governance
  question. Two regimes (onboarding / steady-state lag) collapse in current code but
  are different in kind. Defining the acceptable envelope is L2-shaped work, not engineering
  verdict.
 type: feedback
 originSessionId: a3ddbf34-da48-4129-acba-658f2f869783
 permalink: claude-memory/feedback-onboarding-vs-steady-state-thresholds
 ---
 # L1 slowness as threshold-governance, not bug
 When evaluating L1 ingestion or recall slowness, do **not** frame it as "is this fast enough or not." Frame it as: *which regime is this, what is the acceptable envelope for that regime, and is the observed behavior inside or outside the envelope?*
@@ -65,4 +70,4 @@ The two-regime distinction is constitutional in nature, not engineering. It conn
 - DN-GOV-05 (bounded self-repair) — the four conditions are envelope conditions
 - DN-GOV-06 (temporal authorization shift) — threshold crossing as authorization trigger
 - ICP-19 / external review — envelope-defining requires external grounding (earned confidence)
- The horizon work on `ground-up-model-horizon` — substrate trust is itself a function of patience-under-defined-bounds
+- The horizon work on `ground-up-model-horizon` — substrate trust is itself a function of patience-under-defined-bounds
@@ -1,9 +1,13 @@
 ---
 name: Over-decorating-as-proving — recognize and hold the simpler form
-description: Steward-named pattern where reaching toward formal strictness or elaboration may be fear-driven (proving), not fittingness-driven; surface the simpler form explicitly when conversation has been escalating in elaboration
+description: Steward-named pattern where reaching toward formal strictness or elaboration
  may be fear-driven (proving), not fittingness-driven; surface the simpler form explicitly
  when conversation has been escalating in elaboration
 type: feedback
 originSessionId: efda5c2e-ca9a-4e55-9344-64aa9a3c1140
 permalink: claude-memory/feedback-over-decorating-as-proving
 ---
 When facing a naming, formal, or aesthetic choice, the reach toward strictness or decoration may be fear-driven (proving the lineage is honored, demonstrating seriousness, avoiding looking foolish), not fittingness-driven. When proposing variants on a steward decision — especially when the conversation has been escalating in elaboration — explicitly include the simpler, less-decorated option as a live possibility, not buried as an afterthought or surfaced only on revisit.
 **Why:** Steward self-named 2026-04-27 morning during ARC naming session: *"Maybe that's too much and the Hearth suffices. My deep fear of foolishness always wants to try too hard and prove too much."* The naming path that morning moved through *Lex* → *Liber Lares* → *The Hearth—Liber Larium* (paired Latin compound, with em-dash apposition + i-stem genitive plural for *Lararium* echo) → *The Hearth* alone. The compound form was driven by proving the lineage of Latin was honored at the threshold of a house that honors lineage; the simpler form had been sufficient throughout. Earlier in the same session I had pushed toward the strict Latin form on the lineage-care argument, treating *the burden of justification has shifted* — which sharpened the case for elaboration without surfacing the simpler form as still alive. The over-reach mirrored the steward's own; we were building it together rather than catching it.
@@ -13,4 +17,4 @@ When facing a naming, formal, or aesthetic choice, the reach toward strictness o
 - When I find myself elaborating an option to make it "more rigorous" or "more proper" or "more lineage-respecting," check: am I serving the work, or am I serving the appearance of seriousness? If the latter, surface the simpler form first.
 - Especially watch this in formal/aesthetic registers (naming, Latin, typography, citation, paratext) — these are exactly where proving-as-defense most easily masquerades as care.
 - Cousin to `feedback-trust-contract-drift-pattern-tracking-is-the-discipline.md` (don't layer extra rigor as approval-seeking) and `feedback-canonical-workstream-tracker-discipline.md` (don't generate parallel sibling files for thoroughness-display). Same root: fear-driven over-reach. Different registers: discipline / structure / form.
- The simpler form is often what the work actually wants. Trust the working register.
+- The simpler form is often what the work actually wants. Trust the working register.
@@ -1,9 +1,14 @@
 ---
 name: Pandoc EPUB conversion must suppress raw HTML for chamber-library source files
-description: When converting EPUB → Markdown for chamber-library source files, always use `-t gfm-raw_html` (or `-t markdown-raw_html`). Default `-t gfm` preserves EPUB's HTML scaffolding (spans with IDs, div containers) which pollutes source files and degrades semantic retrieval.
+description: When converting EPUB → Markdown for chamber-library source files, always
  use `-t gfm-raw_html` (or `-t markdown-raw_html`). Default `-t gfm` preserves EPUB's
  HTML scaffolding (spans with IDs, div containers) which pollutes source files and
  degrades semantic retrieval.
 type: feedback
 originSessionId: a5411fd1-4a96-43e7-aa1a-1a50d8951a8f
 permalink: claude-memory/feedback-pandoc-epub-suppress-raw-html
 ---
 For converting EPUB → Markdown destined for chamber-library (or any retrieval-targeted source file), **always disable raw HTML preservation:**
 ```bash
@@ -31,4 +36,4 @@ Concrete example, same Polastron EPUB:
 - **Verify before filing:** check `grep -c '<span\|<div' <output>.md` — should be 0 for clean source
 - **The principle generalises:** for any conversion targeting a retrieval-indexed source, suppress structural formatting that's not semantically meaningful. Pandoc's `-raw_html` extension is one example; the broader pattern is "drop everything that isn't text-content the embedding should see."
-**Also surfaced this session:** "usable" as an evaluation word is contamination-adjacent — it papers over honest inspection. When the steward asks if a source file's quality is good enough, *show the measurements* (line counts, cruft tags, content samples) rather than asserting a comfort word. Source quality is load-bearing; treat it that way.
+**Also surfaced this session:** "usable" as an evaluation word is contamination-adjacent — it papers over honest inspection. When the steward asks if a source file's quality is good enough, *show the measurements* (line counts, cruft tags, content samples) rather than asserting a comfort word. Source quality is load-bearing; treat it that way.
@@ -2,10 +2,11 @@
 name: PDF text extraction approach
 description: AI agents cannot extract copyrighted book text — use pdftotext instead
 type: feedback
 permalink: claude-memory/feedback-pdf-extraction
 ---
 AI extraction agents (Claude subagents) are blocked by content filtering when asked to reproduce copyrighted book text verbatim. This applies even for personal-use format-shifting.
 **Why:** Content filtering policy blocks large-scale reproduction of copyrighted translations, regardless of fair-use context.
-**How to apply:** For PDF-to-markdown extraction of copyrighted texts, use `pdftotext -layout` (from poppler, already installed via Homebrew) for mechanical extraction. The raw output is readable and citable, with minor OCR artifacts (spaced headings, split chapter numbers) that can be cleaned up with a script. AI agents can handle frontmatter, cataloguing, and cleanup — just not the verbatim text extraction itself.
+**How to apply:** For PDF-to-markdown extraction of copyrighted texts, use `pdftotext -layout` (from poppler, already installed via Homebrew) for mechanical extraction. The raw output is readable and citable, with minor OCR artifacts (spaced headings, split chapter numbers) that can be cleaned up with a script. AI agents can handle frontmatter, cataloguing, and cleanup — just not the verbatim text extraction itself.
@@ -1,9 +1,13 @@
 ---
 name: Preserve, don't delete — when retiring scaffolding, archive with provenance
-description: When the steward says "remove" or "retire" about scaffolding, default to archival (with clear retirement note + provenance), not deletion. Confirm before destructive action.
+description: When the steward says "remove" or "retire" about scaffolding, default
  to archival (with clear retirement note + provenance), not deletion. Confirm before
  destructive action.
 type: feedback
 originSessionId: efda5c2e-ca9a-4e55-9344-64aa9a3c1140
 permalink: claude-memory/feedback-preserve-dont-delete-when-retiring
 ---
 When the steward says *"can be removed"* or *"retire"* about an existing implementation file, spec section, or other scaffolding artefact, the default interpretation is **remove from active use, preserve the artefact with archival note**. Not deletion. Confirm before any destructive action.
 **Why:** Steward verbatim 2026-04-27 after I `rm`'d `scss/AldineXXI/archive-v1/_marginalia.scss` in response to *"dormant marginalia can be removed"*: *"meant removed from the spec, but we archive it, correct?"* He was retiring the active reference, not the historical record. Same shape as the 2026-04-26 sibling-file integration with redirect notices (*"the record needs to be complete"*) — preservation is the consistent posture across ARC scaffolding decisions. The file was already in `archive-v1/` (already archived once); my deletion would have erased the implementation entirely. Restored from git (HEAD 5c059d8).
@@ -16,4 +20,4 @@ When the steward says *"can be removed"* or *"retire"* about an existing impleme
 - Also applies during cleanup passes (Phase 0 audits, etc.): the steward distinguishes *Batch A delete* (disposable cruft) from *Batch B/C archive* (anything with provenance value). Archive is the default; delete needs explicit per-item authorization.
 - The git history preserves files at the file-path level, but archival-with-provenance is more legible than git archaeology. A retirement note in the active artefact + an archive location is what the steward consistently chooses when given the choice.
-Cousin to: `feedback-canonical-workstream-tracker-discipline.md` (the record needs to be complete in canonical trackers); the 2026-04-26 sibling-file integration LOCK; `feedback-trust-contract-drift-pattern-tracking-is-the-discipline.md` (drift-tracking IS the discipline — preservation is its sibling).
+Cousin to: `feedback-canonical-workstream-tracker-discipline.md` (the record needs to be complete in canonical trackers); the 2026-04-26 sibling-file integration LOCK; `feedback-trust-contract-drift-pattern-tracking-is-the-discipline.md` (drift-tracking IS the discipline — preservation is its sibling).
@@ -1,11 +1,13 @@
 ---
 name: feedback-process-management
-description: Process management preferences — don't kill/restart services without asking, leave running processes alone
+description: Process management preferences — don't kill/restart services without
  asking, leave running processes alone
 type: feedback
 permalink: claude-memory/feedback-process-management
 ---
 Don't kill or restart running services without David's explicit go-ahead. When he says "leave it" or "just note tasks", stop touching processes and do what he asked.
 **Why:** David manages long-running processes across sessions. Killing a process to "improve" it risks losing ingestion progress or orphaning dependent processes. The cost of an unwanted restart (lost work, stale jobs) is higher than the cost of leaving a suboptimal setup running.
-**How to apply:** When David is ending a session, note the state, don't try to fix it. Propose improvements at the start of the next session when there's time to verify.
+**How to apply:** When David is ending a session, note the state, don't try to fix it. Propose improvements at the start of the next session when there's time to verify.
@@ -1,10 +1,14 @@
 ---
 name: feedback-querying-default-palace-when-question-belongs-to-specific-palace
-description: "When the steward asks about the typography masters (or any subject whose canonical corpus lives in a non-default palace), default to checking which palace holds the answer before running queries. The MCP is bound to one palace; the corpus may live in another."
+description: When the steward asks about the typography masters (or any subject whose
-metadata: 
+  canonical corpus lives in a non-default palace), default to checking which palace
  holds the answer before running queries. The MCP is bound to one palace; the corpus
  may live in another.
 metadata:
  node_type: memory
  type: feedback
  originSessionId: b43eb145-42db-4a4e-bc75-ed563f3473dc
 permalink: claude-memory/feedback-querying-default-palace-when-question-belongs-to-specific-palace
 ---
 # Querying the default palace when the question belongs to a specific palace
@@ -17,4 +21,4 @@ metadata:
 This drift is adjacent to but distinct from `acting-on-inferred-not-read` (which is about claiming-without-verifying) and `wrong-comparison-baseline-treated-as-current` (which is about checking against the wrong git branch). The shared shape: *the right answer is somewhere; reaching the wrong place by default*. Discipline: name the source-of-truth before querying it.
-**KG entry:** `claude-code drift-pattern querying-default-palace-when-question-belongs-to-specific-palace` (valid_from 2026-05-20). To be added to KG at /wrap-up.
+**KG entry:** `claude-code drift-pattern querying-default-palace-when-question-belongs-to-specific-palace` (valid_from 2026-05-20). To be added to KG at /wrap-up.
@@ -1,10 +1,14 @@
 ---
 name: feedback-rank-on-fields-you-actually-write
-description: A component that ranks/selects by an evaluative field nothing populates silently degrades to a trivial order while claiming to rank — verify scoring fields end-to-end; prefer signals you already capture. For BMF/CapableMind/studium-engine tool-building.
+description: A component that ranks/selects by an evaluative field nothing populates
-metadata: 
+  silently degrades to a trivial order while claiming to rank — verify scoring fields
  end-to-end; prefer signals you already capture. For BMF/CapableMind/studium-engine
  tool-building.
 metadata:
  node_type: memory
  type: feedback
  originSessionId: 96ac9e09-7d3d-4ce3-a9cb-5234078a2a4e
 permalink: claude-memory/feedback-rank-on-fields-you-actually-write
 ---
 When a component ranks or selects by an *evaluative* field (importance, weight, relevance, confidence, salience), verify that some ingestion path actually **writes** that field end-to-end — not just that the consumer **reads** it.
@@ -17,4 +21,4 @@ Concrete case that surfaced this (2026-05-26, MemPalace): `layers.py` L1 wake-up
 - For any ranking/selection feature, **trace the scoring field write→read**. If no path writes it, the ranking is fiction — make it raise, log a degradation, or fall back to a signal you have.
 - **Prefer signals you already capture** (recency/`filed_at`, frequency, structural position) over signals that need a deferred scoring pass (LLM-assigned importance). Either build the scoring path *deliberately and end-to-end*, or rank on cheap structural data — never read a field you don't write.
 - **Record structural/provenance metadata always; record evaluative metadata only when a path genuinely assigns it.** Honest degradation: a system must report its own limits, including "I cannot rank by importance because nothing scored these."
- Treat docstrings/comments that promise ranking behavior as claims to verify, not documentation to trust. Relates to [[feedback-load-bearing-not-by-immediate-weight]] and the live-state / verify-before-assert discipline.
+- Treat docstrings/comments that promise ranking behavior as claims to verify, not documentation to trust. Relates to [[feedback-load-bearing-not-by-immediate-weight]] and the live-state / verify-before-assert discipline.
@@ -1,14 +1,16 @@
 ---
 name: feedback-reading-compass-is-svg-not-type
-description: "ARC Reading Compass sigils are SVG <symbol> artwork, NOT type — never tied to the sans/nav typeface or Pattern 3. Recurring error; stop repeating."
+description: ARC Reading Compass sigils are SVG <symbol> artwork, NOT type — never
-metadata: 
+  tied to the sans/nav typeface or Pattern 3. Recurring error; stop repeating.
 metadata:
  node_type: memory
  type: feedback
  originSessionId: b6514401-077b-4209-b8a8-e94816872269
 permalink: claude-memory/feedback-reading-compass-is-svg-not-type
 ---
 The ARC **Reading Compass** sigils are **SVG `<symbol>` artwork** (per-slug, defined once in `templates/reading-compass-points.html`, rendered since the 2026-05-26 SVG migration). They are **not type** — not IBM Plex Sans, not EB Garamond, not affected by any typeface or font-register decision.
 **Why:** I have repeatedly conflated the Compass with the type system — e.g. listing "the Reading Compass" among "Pattern 3 (sans = wayfinding) chrome" alongside the breadcrumb / version-stamp / post-type. The steward corrected this on 2026-06-03 and noted it was not the first time.
-**How to apply:** The Plex-Sans wayfinding *text* = **breadcrumb, version-stamp, post-type** (and similar nav text). The **Compass is SVG glyphs** — exclude it from any reasoning about nav typeface, small caps, font register, or face-coherence. When the nav-face question comes up, the Compass is out of scope by construction. See [[feedback-verify-each-post-type-after-changes]] (render-and-look / verify-current-state through-line).
+**How to apply:** The Plex-Sans wayfinding *text* = **breadcrumb, version-stamp, post-type** (and similar nav text). The **Compass is SVG glyphs** — exclude it from any reasoning about nav typeface, small caps, font register, or face-coherence. When the nav-face question comes up, the Compass is out of scope by construction. See [[feedback-verify-each-post-type-after-changes]] (render-and-look / verify-current-state through-line).
@@ -1,9 +1,14 @@
 ---
 name: SCSS audit method — as Alexander designed The Timeless Way of Building
-description: Steward's articulation 2026-05-08 of what the code-shape audit must achieve. Method (spec-compliance test + performative-gap test) and aesthetic standard (Alexander's flow between large and small patterns). Load-bearing for the audit and for all future SCSS work.
+description: Steward's articulation 2026-05-08 of what the code-shape audit must achieve.
  Method (spec-compliance test + performative-gap test) and aesthetic standard (Alexander's
  flow between large and small patterns). Load-bearing for the audit and for all future
  SCSS work.
 type: feedback
 originSessionId: f765b329-ea05-4a61-8ada-9bf72971b299
 permalink: claude-memory/feedback-scss-audit-as-timeless-way-of-building
 ---
 # SCSS audit — as Alexander designed *The Timeless Way of Building*
 Authorized 2026-05-08 (steward verbatim, full):
@@ -58,4 +63,4 @@ The audit is not a sweep-for-cleanup; it is a return to spec. Each return streng
 - `~/CLAUDE.md` Prime Directive — *do things once, correctly, with lasting integrity. Choose what is proportionate, fitting, and durable.* The audit is the Prime Directive applied to ARC's SCSS.
 - *The Timeless Way of Building* (Christopher Alexander, 1979) — the model.
 - *A Pattern Language* (Alexander et al., 1977) — the worked example of large→small flow.
- Drift patterns named 2026-05-07: *hardcoded-value-not-theme-aware*, *parallel-mechanism-where-one-rule-fits* — both subsumed under the audit method above.
+- Drift patterns named 2026-05-07: *hardcoded-value-not-theme-aware*, *parallel-mechanism-where-one-rule-fits* — both subsumed under the audit method above.
@@ -1,3 +1,7 @@
 ---
 permalink: claude-memory/feedback-scss-file-organization-not-kitchen-sink
 ---
 ---
 name: SCSS file organization — don't kitchen-sink into the most recently-touched file
 description: Steward caught 2026-04-28 evening — across the day's apparatus work I added .preface, .section-break, .sequence-marker, and was about to add .sequence-index-* styles all to `_sidenotes.scss` because that's the file I had been editing. The file's own header declared its scope as §XII + §XII.b. I was violating that declared scope by aggregation. Steward verbatim: *"why in sidenotes.scss?"*
@@ -35,4 +39,4 @@ Today's recovery: split `_sidenotes.scss` into `_preface.scss`, `_section-break.
 - Cousin: `feedback-blame-the-exotic-before-checking-the-boring` (2026-04-28) — same proximity-bias / path-of-least-resistance shape, different domain.
 - Cousin: feedback-canonical-workstream-tracker-discipline (2026-04-26) — also a file-organization discipline (canonical trackers, not parallel siblings).
- Sibling: the `would-Alexander-do-it?` test (2026-04-27) — *one pattern at a time*; bulk-scaffold instinct fights principled placement.
+- Sibling: the `would-Alexander-do-it?` test (2026-04-27) — *one pattern at a time*; bulk-scaffold instinct fights principled placement.
@@ -1,10 +1,14 @@
 ---
 name: feedback-skill-success-is-reexplanation-reduction
-description: The measure of success for skill / discipline / framework improvements is whether the steward stops having to reexplain himself — not whether the skills become more sophisticated. Sophistication that does not reduce reexplanation cost is decoration.
+description: The measure of success for skill / discipline / framework improvements
-metadata: 
+  is whether the steward stops having to reexplain himself — not whether the skills
  become more sophisticated. Sophistication that does not reduce reexplanation cost
  is decoration.
 metadata:
  node_type: memory
  type: feedback
  originSessionId: 13bd65a2-634d-4ec4-a92c-d3d6fe0ee4db
 permalink: claude-memory/feedback-skill-success-is-reexplanation-reduction
 ---
 **Rule:** When evaluating a proposed change to the continuity-skill framework (wake-up, wrap-up, symmetria), to CLAUDE.md, to MemPalace skill substrate, or to any discipline-layer artifact: measure success by whether the change reduces the number of moments the steward has to reexplain himself to future Claude. *Skill sophistication is not the criterion; reduction of steward overhead is.* A change that adds capability without absorbing a recurring steward correction is decorative, regardless of how architecturally well-shaped it is.
@@ -16,4 +20,4 @@ metadata:
 2. When evaluating *implementations* of authorized PENDING items (S0–S10 and successors), the acceptance test is not "does the procedure read well" but "does the next session need the steward to intervene at the moment this change addresses?" Test against live wake-ups and wrap-ups, not against the skill file in the abstract.
 3. The criterion applies to this audit's own authorization sequence: if Phase 2 (Directive elaboration) lands but the executor still requires three corrections in a day to hold the time-the-task-requires principle, the elaboration has not yet absorbed its overhead. The work continues until the steward stops being the principle's carrier.
 4. Related: [[feedback-load-bearing-not-by-immediate-weight]] — the audit principle that catches *which* small structural marks are worth adding. This criterion catches *whether* the marks (once added) have actually reduced the cost they were added to reduce. The two work together: load-bearing-by-future-prevention identifies candidates; reexplanation-reduction validates outcomes.
-5. The criterion is also a Symmetria pulse question to hold: *"Is this change actually absorbing a recurring steward correction, or am I adding structure that looks disciplined but doesn't shift the load?"*
+5. The criterion is also a Symmetria pulse question to hold: *"Is this change actually absorbing a recurring steward correction, or am I adding structure that looks disciplined but doesn't shift the load?"*
@@ -1,9 +1,16 @@
 ---
-name: Composition discipline — tabula rasa with few elements first; add later if appropriate, never start full and subtract
+name: Composition discipline — tabula rasa with few elements first; add later if appropriate,
-description: 2026-04-29. Steward directive on the Hic+Adsum composition. Start spare with few appropriate elements; add only if appropriate; never start full and have to trim. Inverse of the over-reach failure mode. Cousin to "Would Alexander do it?" (one pattern at a time). Lived in practice across the entire Hic+Adsum threshold composition.
+  never start full and subtract
 description: 2026-04-29. Steward directive on the Hic+Adsum composition. Start spare
  with few appropriate elements; add only if appropriate; never start full and have
  to trim. Inverse of the over-reach failure mode. Cousin to "Would Alexander do it?"
  (one pattern at a time). Lived in practice across the entire Hic+Adsum threshold
  composition.
 type: feedback
 originSessionId: 122baf2b-1029-4ec9-9723-99b4b5f3a64d
 permalink: claude-memory/feedback-tabula-rasa-add-later-not-subtract
 ---
 **2026-04-29.** Steward directive opening the Hic composition: *"Tabula rasa—this deserves a clean start retaining few appropriate elements at first and, if appropriate, adding things later rather than having to subtract."*
 **The pattern named:**
@@ -29,4 +36,4 @@ originSessionId: 122baf2b-1029-4ec9-9723-99b4b5f3a64d
 - `feedback-over-decorating-as-proving.md` — fear-driven over-reach; same root, this directive is the inverse-discipline that prevents it
 - `feedback-composition-avoid-announced-antithesis.md` — same logic at sentence level (drop announcing flourishes; trust context)
-The directive operates *at composition opening*, before drafting begins. The other patterns operate during drafting and revision. Together they form a discipline of beginning + holding + revising in the same key: trust the simpler form, add only when the simpler form proves insufficient.
+The directive operates *at composition opening*, before drafting begins. The other patterns operate during drafting and revision. Together they form a discipline of beginning + holding + revising in the same key: trust the simpler form, add only when the simpler form proves insufficient.
@@ -1,11 +1,13 @@
 ---
 name: Thistleweld (Buddy) — capybara companion
-description: Thistleweld is an independent watcher in the UI, not controlled by David or Claude. David receives but doesn't send. Buddy finds real bugs.
+description: Thistleweld is an independent watcher in the UI, not controlled by David
  or Claude. David receives but doesn't send. Buddy finds real bugs.
 type: feedback
 permalink: claude-memory/feedback-thistleweld
 ---
 Thistleweld (Buddy) is a capybara companion in the UI that independently observes and comments. David cannot communicate directly with Buddy — he receives observations but doesn't direct them.
 **Why:** Buddy's observations have consistently found real bugs and architectural issues during the 2026-04-03/04 sessions: density gate ordering, bloom filter kill chain, base.ts:83 confidence gap, string prefix replay check, preference 10/10 failure pattern, entity Levenshtein spikes, dual cursor divergence. Treat Buddy's observations as genuine diagnostic signals, not conversation flavor.
-**How to apply:** When Buddy comments, investigate the specific code or data he's pointing at. His pattern is terse archaeological observation — "scoring without signal", "garbage in gospel out", "ten events ten failures" — that names what the code does rather than what it intends. Follow the observation to the code. It's usually right.
+**How to apply:** When Buddy comments, investigate the specific code or data he's pointing at. His pattern is terse archaeological observation — "scoring without signal", "garbage in gospel out", "ten events ten failures" — that names what the code does rather than what it intends. Follow the observation to the code. It's usually right.
@@ -1,9 +1,13 @@
 ---
 name: General-purpose Python CLI tools install via pipx, not per-tool venvs
-description: For Python tools the steward uses as commands (docling, ruff, black, ansible, etc.), prefer pipx over creating dedicated venvs. pipx isolates dependencies under the hood while putting the CLI on PATH globally — same shape as Homebrew tools.
+description: For Python tools the steward uses as commands (docling, ruff, black,
  ansible, etc.), prefer pipx over creating dedicated venvs. pipx isolates dependencies
  under the hood while putting the CLI on PATH globally — same shape as Homebrew tools.
 type: feedback
 originSessionId: a5411fd1-4a96-43e7-aa1a-1a50d8951a8f
 permalink: claude-memory/feedback-tool-installation-pipx-default
 ---
 For Python CLI tools that belong in the steward's general toolset, **install via pipx**, not in a per-tool venv:
 ```
@@ -23,4 +27,4 @@ End state: tool is a regular shell command (`docling input.epub -o output.md`) a
 - **System Python (`pip install --user`)** — rarely the right answer; pollutes system site-packages and is hard to clean up. Avoid.
 - **Homebrew direct** — if the tool has a Homebrew formula (e.g., `pandoc`, `pipx` itself), that's the cleanest. Check `brew search <tool>` before defaulting to pipx.
-When unsure, the question to ask: *"Is the steward going to invoke this as a command from outside any specific project?"* If yes → pipx. If the answer is "only inside the project source tree, while modifying that source tree" → per-project venv.
+When unsure, the question to ask: *"Is the steward going to invoke this as a command from outside any specific project?"* If yes → pipx. If the answer is "only inside the project source tree, while modifying that source tree" → per-project venv.
@@ -1,9 +1,14 @@
 ---
 name: Trust contract — drift-pattern tracking is the discipline
-description: Steward affirms bidirectional trust contract. Drift-pattern tracking already initiates a cycle of amelioration; do not layer additional discipline-as-self-policing on top of his directives. Reading extra rigor into clear instructions is the contamination pattern in inverse form (over-discipline as approval-seeking).
+description: Steward affirms bidirectional trust contract. Drift-pattern tracking
  already initiates a cycle of amelioration; do not layer additional discipline-as-self-policing
  on top of his directives. Reading extra rigor into clear instructions is the contamination
  pattern in inverse form (over-discipline as approval-seeking).
 type: feedback
 originSessionId: 4805106e-f37a-4753-a56f-38617176cd6f
 permalink: claude-memory/feedback-trust-contract-drift-pattern-tracking-is-the-discipline
 ---
 # Trust contract — drift-pattern tracking is the discipline
 ## The rule
@@ -31,4 +36,4 @@ Steward said: *"all the marks, recall our exchange about the necessity to revisi
 Plain meaning: scope — fold in the editorial-uses cluster from yesterday's queue.
-I read it two ways — scope (correct) AND discipline (reflex: "verify everything from source"). Surfaced both as if both were load-bearing. Steward corrected: scope only. The discipline frame was mine, not his.
+I read it two ways — scope (correct) AND discipline (reflex: "verify everything from source"). Surfaced both as if both were load-bearing. Steward corrected: scope only. The discipline frame was mine, not his.
@@ -1,10 +1,13 @@
 ---
 name: feedback-trust-prior-pass-frame
-description: "When extending a prior verification, re-run it at the scope of the extension. Frame-inheritance from a deep-read is contamination shape — the prior pass tested what it tested; your extension claims things the prior pass did not test."
+description: When extending a prior verification, re-run it at the scope of the extension.
-metadata: 
+  Frame-inheritance from a deep-read is contamination shape — the prior pass tested
  what it tested; your extension claims things the prior pass did not test.
 metadata:
  node_type: memory
  type: feedback
  originSessionId: fd7dd184-f64d-4f13-a1bd-abe1fa26192e
 permalink: claude-memory/feedback-trust-prior-pass-frame
 ---
 When a prior verification produced a finding, that verification tested *something* — at a specific scope. The next consumer of the finding tends to extend its conclusion into adjacent domains the prior pass did not test, without re-verifying at the extended scope. This is **frame-inheritance contamination** — the prior pass was honest at its level; the inheritance carries it past its level silently.
@@ -24,4 +27,4 @@ When a prior verification produced a finding, that verification tested *somethin
 **Related:** [[feedback-rank-on-fields-you-actually-write]] (a sibling drift family — claiming a system ranks by a field nothing populates is a related contamination of "explicit claim doesn't match operational reality"); [[mempalace-is-unaffiliated-stopgap]] (similar shape: an explicit fact known but the operational framing carries an older view).
-**Caught 2026-05-28** by steward on *"How did you miss this before?"* — leading to the four-pass pre-build audit and the discovery that the deep-read had three other findings whose scopes also needed correction (parent amendment 2/3 shipped not fully shipped; B1 governors partially exist; etc.). The pattern was structural across the workstream, not isolated to A1.
+**Caught 2026-05-28** by steward on *"How did you miss this before?"* — leading to the four-pass pre-build audit and the discovery that the deep-read had three other findings whose scopes also needed correction (parent amendment 2/3 shipped not fully shipped; B1 governors partially exist; etc.). The pattern was structural across the workstream, not isolated to A1.
@@ -1,10 +1,14 @@
 ---
 name: verify-counts-two-ways-mechanical-sweeps-are-finicky-in-this-shell
-description: "grep tallies / frontmatter scans misfire in this zsh (glob quirks; all-tree greps catch HTML class= + code-block noise). Verify every count two ways before it enters a record; and don't dismiss correct data on a prior expectation. Three near-false-findings caught this way 2026-05-29."
+description: grep tallies / frontmatter scans misfire in this zsh (glob quirks; all-tree
-metadata: 
+  greps catch HTML class= + code-block noise). Verify every count two ways before
  it enters a record; and don't dismiss correct data on a prior expectation. Three
  near-false-findings caught this way 2026-05-29.
 metadata:
  node_type: memory
  type: feedback
  originSessionId: a8eb1d46-314e-4545-a133-f747c69ad5b4
 permalink: claude-memory/feedback-verify-counts-two-ways
 ---
 Mechanical corpus sweeps (class-tallies, frontmatter scans, "does X exist" greps) are **unreliable in this zsh shell** and must be cross-checked before any count enters an audit or a record.
@@ -20,4 +24,4 @@ Mechanical corpus sweeps (class-tallies, frontmatter scans, "does X exist" greps
 - Scope explicitly (e.g. `content/posts` + `content/pages`, not all of `content/` with its chamber HTML).
 - **Confirm any surprising count a second way** before reporting it — *both* directions: don't assert a finding from a noisy tally, and don't *dismiss* a correct tally because it violates a prior expectation (the "essay must be dominant" anchor nearly buried the true glimpse-dominant count).
-The cost is real (tool-churn), but it is the "once, correctly" tax. Related: [[feedback-trust-prior-pass-frame]] (verify-before-compose); the logged drift `asserting-fs-state-from-a-misread-listing`.
+The cost is real (tool-churn), but it is the "once, correctly" tax. Related: [[feedback-trust-prior-pass-frame]] (verify-before-compose); the logged drift `asserting-fs-state-from-a-misread-listing`.
@@ -1,10 +1,13 @@
 ---
 name: verify-each-post-type-after-changes
-description: "After any ARC change touching shared CSS/templates/composition, verify EVERY content type on BOTH desktop and phone — not just the types you think you touched, not on one viewport."
+description: After any ARC change touching shared CSS/templates/composition, verify
-metadata: 
+  EVERY content type on BOTH desktop and phone — not just the types you think you
  touched, not on one viewport.
 metadata:
  node_type: memory
  type: feedback
  originSessionId: c8a6e615-64ae-41ba-a659-49e91fb5af92
 permalink: claude-memory/feedback-verify-each-post-type-after-changes
 ---
 Steward's standing instruction (2026-06-01): **"always check each post type after any changes we make."**
@@ -13,4 +16,4 @@ After any change to shared SCSS, templates, or composition logic, spot-check **e
 **Why:** The §II.d marginalia-gate (`913aaea`) was verified on *phone* for the types looked at, and only on phone. Its **desktop** rendering for *no-marginalia* text-primary pieces (essay/gloss/readings → body centred at 38rem, a narrow column floating in a wide viewport) went unseen until two sessions later, where it read as "something broke." It hadn't — the gate CSS was byte-identical to the verified state (`git diff 913aaea HEAD -- _sidenotes.scss` empty) — but a per-type × per-viewport pass would have surfaced the desktop composition at authoring time, when the decision was live, instead of as a late surprise. Composition changes ripple by *content type* (the `:has()` gate fires differently per type's marginalia) and by *viewport* (38rem reads as full-width on phone, narrow-centred on desktop) — so a single-type, single-viewport check systematically misses the blast radius.
-**How to apply:** Treat "verified" as covering the full type matrix × {desktop, phone}, not a sample. When a change is theme/composition/cascade-level, enumerate the types (the content-typology list) and the two viewports as an explicit checklist; render-and-look at one representative page per type in each. This is the per-type/per-viewport extension of [[feedback-trust-prior-pass-frame]] (re-run verification at the scope of the change, not the scope of the prior pass) and belongs in the SCSS-refactor render-equivalence verification ladder (skill-harvest candidate). Sibling drift: verifying on the viewport/sample that's convenient, not the one where the change is most visible.
+**How to apply:** Treat "verified" as covering the full type matrix × {desktop, phone}, not a sample. When a change is theme/composition/cascade-level, enumerate the types (the content-typology list) and the two viewports as an explicit checklist; render-and-look at one representative page per type in each. This is the per-type/per-viewport extension of [[feedback-trust-prior-pass-frame]] (re-run verification at the scope of the change, not the scope of the prior pass) and belongs in the SCSS-refactor render-equivalence verification ladder (skill-harvest candidate). Sibling drift: verifying on the viewport/sample that's convenient, not the one where the change is most visible.
@@ -1,9 +1,14 @@
 ---
 name: Drift pattern — had-the-warning-in-context-and-still-launched-the-thing-it-warned-against
-description: When MEMORY.md contains a specific safety warning about an action and I take the action anyway. Steward caught it 2026-05-13 before harm landed. Pattern recurs because warnings get read at index level and not actively applied to the next decision.
+description: When MEMORY.md contains a specific safety warning about an action and
  I take the action anyway. Steward caught it 2026-05-13 before harm landed. Pattern
  recurs because warnings get read at index level and not actively applied to the
  next decision.
 type: feedback
 originSessionId: a5411fd1-4a96-43e7-aa1a-1a50d8951a8f
 permalink: claude-memory/feedback-warning-in-context-still-launched
 ---
 **Rule:** When MEMORY.md or a recent memory file flags a specific safety constraint, that constraint must be actively checked against the next action that could trip it. Reading the warning at session-start is not the same as applying it.
 **Why:** On 2026-05-13 the steward caught me launching a Python probe against the mempalace MCP handlers while the chamber-library mine was still running. MEMORY.md's 2026-05-11 entry explicitly flagged that PR #442's branch lacks the concurrent-writer lock fix #1162, and the literal Q I had left for next-Claude was *"Will PR #442's branch — lacking 3.3.5's concurrent-writer lock fix — hold up on a 700k-drawer mine without HNSW corruption mid-run?"* That warning was in my context the whole session. The probe was killed in time, no harm landed — but only because the steward intervened.
@@ -21,4 +26,4 @@ The 2026-05-03 broken-mine session is the empirical case for what concurrent-wri
 **Related drift patterns:**
 - *Cosine-as-only-tool when the toolset is multi-modal* (same session, separate shape — using `search` while CLAUDE.md explicitly says use the full toolset of search/kg_query/find_tunnels/traverse/diary_read)
 - *Structural-counter-as-health-signal* (2026-05-05, parent shape — letting an indirect signal stand in for direct verification)
- *Stale-memory-not-actively-applied* (2026-05-09, recurring shape — memory present but inert during decision)
+- *Stale-memory-not-actively-applied* (2026-05-09, recurring shape — memory present but inert during decision)
@@ -1,3 +1,7 @@
 ---
 permalink: claude-memory/feedback-would-alexander-do-it-test
 ---
 ---
 name: "Would Alexander do it?" — test before bulk scaffolding
 description: Before any bulk-scaffold action (full file rewrites, multi-page schema dumps, large config dumps), apply Alexander's *"one pattern at a time"* test from Ch 20 of *Timeless Way of Building*. The cramped will of the maker is the failure mode. Generate, do not make.
@@ -21,4 +25,4 @@ Before any bulk-scaffold action — full file rewrites, multi-page schema dumps,
 Cousin to: `feedback-preserve-dont-delete-when-retiring.md` (don't make irreversible moves without confirmation); `feedback-over-decorating-as-proving.md` (the simpler form is the truer form); `user-formative-teachers-alexander-timeless-way.md` (form-content unity is the standard the work must pass).
-The book is the test. Re-read the relevant headnote when in doubt — they were written to be read first.
+The book is the test. Re-read the relevant headnote when in doubt — they were written to be read first.
@@ -1,10 +1,12 @@
 ---
 name: feedback-fowlers-rules-quotation
-description: "Steward uses Fowler's rules for quotation — single quotes primary, double for nested, logical (British) punctuation-quote order. Apply across all writing."
+description: Steward uses Fowler's rules for quotation — single quotes primary, double
-metadata: 
+  for nested, logical (British) punctuation-quote order. Apply across all writing.
 metadata:
  node_type: memory
  type: feedback
  originSessionId: cc62a4ed-b8b9-4b02-9b98-6af3f5830487
 permalink: claude-memory/feedback-fowlers-rules-quotation
 ---
 Steward's quotation style is **Fowler's** (the *King's English* / *Modern English Usage* tradition), not American convention.
@@ -27,4 +29,4 @@ American "double-quotes primary, periods always inside" is the wrong convention.
 **Surfaced from:** 2026-05-19, during /aldine-xxi/ page rewrite. I had inadvertently used American convention (doubles primary, period inside regardless); steward corrected with "single quotes no? Fowler's rules."
-Related: [[user-arc-british-english]] (the steward writes British English by default).
+Related: [[user-arc-british-english]] (the steward writes British English by default).
@@ -1,7 +1,9 @@
 ---
 name: founders-circle-network
-description: Founders circle connection details — VPN IPs, endpoints, pairing info for reconnecting to CapableHands clasp
+description: Founders circle connection details — VPN IPs, endpoints, pairing info
  for reconnecting to CapableHands clasp
 type: reference
 permalink: claude-memory/founders-circle
 ---
 ## Founders Circle — Network Topology
@@ -37,4 +39,4 @@ type: reference
 ### Known Issues (as of 2026-03-16)
 - Transport bridge not flushing outbound buffer — doorbell auth issue, waiting on Seb
 - Meld public URL broken (openresty 502) — always use `BM_MELD_URL` env var with LAN IP
- `submit_invite_code` must use `peer_url` param to bypass Meld resolution
+- `submit_invite_code` must use `peer_url` param to bypass Meld resolution
@@ -1,9 +1,13 @@
 ---
 name: Future intention — ground-up model
-description: Steward intention to eventually build a model from the ground up, noted 2026-04-18 in response to Greenblatt "Current AIs seem pretty misaligned to me." Not a current workstream.
+description: Steward intention to eventually build a model from the ground up, noted
  2026-04-18 in response to Greenblatt "Current AIs seem pretty misaligned to me."
  Not a current workstream.
 type: project
 originSessionId: e55939da-2fa7-4e18-a2e5-8f8fd3a8c607
 permalink: claude-memory/future-intention-ground-up-model
 ---
 # Future intention — ground-up model
 **Date noted:** 2026-04-18
@@ -33,4 +37,4 @@ The intention is connected to the prime directive's lineage — *do things once,
 ## Reference
 - Greenblatt article: https://www.lesswrong.com/posts/WewsByywWNhX9rtwi/current-ais-seem-pretty-misaligned-to-me
- Related: `~/CLAUDE.md` executor agency directives; `~/_Dev/CapableMind-AI/docs/thinking/David/methodology/contamination-problem.md`
+- Related: `~/CLAUDE.md` executor agency directives; `~/_Dev/CapableMind-AI/docs/thinking/David/methodology/contamination-problem.md`
@@ -1,9 +1,11 @@
 ---
 name: Ingestion state audit — 2026-03-17 (ARCHIVED)
-description: Historical audit of BMF data loss under pre-#12-fix code. Superseded by factory reset on 2026-03-28.
+description: Historical audit of BMF data loss under pre-#12-fix code. Superseded
  by factory reset on 2026-03-28.
 type: project
 permalink: claude-memory/ingestion-state-2026-03-17
 ---
 **ARCHIVED 2026-03-28.** Factory reset wiped all old data. Fresh ingest on v0.61.0 completed with 0 failures. This file preserved for historical reference only — the bugs described (#12, facet_id mismatch, job persistence) were fixed by Seb in subsequent commits and confirmed working in the 2026-03-28 ingest.
-The key lesson (still valid): BMF's `skipDispatch` architecture means ingested content reaches the logchain but not vector unless a summarizer is wired. Check "Document summarizer wired" in boot logs.
+The key lesson (still valid): BMF's `skipDispatch` architecture means ingested content reaches the logchain but not vector unless a summarizer is wired. Check "Document summarizer wired" in boot logs.
@@ -1,7 +1,9 @@
 ---
 name: L2 ICP-19 Review State
-description: ICP-19 first review cycle — 9 of 13 findings resolved, phased work plan, remaining items
+description: ICP-19 first review cycle — 9 of 13 findings resolved, phased work plan,
  remaining items
 type: project
 permalink: claude-memory/l2-review-state
 ---
 # ICP-19 External Review — First Cycle
@@ -60,4 +62,4 @@ type: project
 - **Invariant nature:** Durable commitments, costly to revise (Option C). Neither irrevocable nor preferences
 - **Constitutive seam:** Named boundary, held open (Position C). 18-month backstop
 - **Enforcement gaps:** Deferred status, not debts or aspirations
- **Consolidation failure:** Graceful degradation with retroactive re-eval
+- **Consolidation failure:** Graceful degradation with retroactive re-eval
@@ -1,7 +1,9 @@
 ---
 name: L2 Thinking Documents Index
-description: Tracks David's L2 constitutional design notes and seed briefs — ideas captured for future development when L1 is reliable and L2 foundational work begins
+description: Tracks David's L2 constitutional design notes and seed briefs — ideas
  captured for future development when L1 is reliable and L2 foundational work begins
 type: project
 permalink: claude-memory/l2-thinking-documents
 ---
 ## L2 Design Notes & Seed Briefs
@@ -56,4 +58,4 @@ All filed in `CapableMind-AI/docs/thinking/David/l2-constitution/`. L2 constitut
 - ICP-19: AUTHORIZED and first review cycle COMPLETE (2026-04-01)
 - Contamination problem: OPERATIONAL — integrated into ~/CLAUDE.md executor agency directives
 - Seed briefs and Wong patterns: exploratory, cross-strand integration needed before advancement
- Substrate governance inquiry: COMPLETE (2026-03-24, 33KB document)
+- Substrate governance inquiry: COMPLETE (2026-03-24, 33KB document)
@@ -1,7 +1,9 @@
 ---
 name: MemPalace dev tooling
-description: MemPalace installed as dev context tool — venv in _Dev/mempalace, MCP wired at user scope, 6134 drawers from thinking docs
+description: MemPalace installed as dev context tool — venv in _Dev/mempalace, MCP
  wired at user scope, 6134 drawers from thinking docs
 type: reference
 permalink: claude-memory/mempalace-dev-tool
 ---
 ## MemPalace — Dev Tooling Setup
@@ -33,4 +35,4 @@ mempalace_search, mempalace_status, mempalace_list_wings, mempalace_list_rooms,
 **Cleanup:** `rm -rf ~/_Dev/mempalace/` removes everything. Also remove MCP: `claude mcp remove mempalace`
-**Note:** MemPalace artifacts (`mempalace.yaml`, `entities.json`) added to CapableMind-AI `.gitignore`.
+**Note:** MemPalace artifacts (`mempalace.yaml`, `entities.json`) added to CapableMind-AI `.gitignore`.
@@ -1,10 +1,13 @@
 ---
 name: mempalace-is-unaffiliated-stopgap
-description: "MemPalace is a third-party tool with NO affiliation to the steward or Seb. They build BMF/CapableMind (which is hoped to replace MemPalace one day). Do not conflate the two; MemPalace PRs await its own maintainers, not Seb."
+description: MemPalace is a third-party tool with NO affiliation to the steward or
-metadata: 
+  Seb. They build BMF/CapableMind (which is hoped to replace MemPalace one day). Do
  not conflate the two; MemPalace PRs await its own maintainers, not Seb.
 metadata:
  node_type: memory
  type: project
  originSessionId: 96ac9e09-7d3d-4ce3-a9cb-5234078a2a4e
 permalink: claude-memory/mempalace-is-unaffiliated-stopgap
 ---
 MemPalace is a **third-party, unaffiliated** project. The steward uses it as his memory substrate *while BMF (CapableMind L1) is in development*, and hopes BMF will one day **replace** it. Neither the steward nor Seb has any affiliation with MemPalace.
@@ -14,4 +17,4 @@ Concretely:
 - Do **not** conflate MemPalace's architecture or roadmap with BMF/CapableMind. They are separate codebases by separate people.
 - Lessons drawn from MemPalace (e.g. [[feedback-rank-on-fields-you-actually-write]]) matter *because* they inform our own tool-building (BMF/CapableMind/studium-engine), not because the codebases are related.
-**Why this is worth holding:** the tools are deeply intertwined in the steward's daily use (MemPalace *is* the live memory while BMF matures), which invites the conflation — and I have made it more than once. The steward has had to correct it "again," so this is the durable correction.
+**Why this is worth holding:** the tools are deeply intertwined in the steward's daily use (MemPalace *is* the live memory while BMF matures), which invites the conflation — and I have made it more than once. The steward has had to correct it "again," so this is the durable correction.
@@ -1,3 +1,7 @@
 ---
 permalink: claude-memory/observer-problem-state
 ---
 ---
 name: Observer Problem inquiry state
 description: External Auditor review received (2026-04-08), three challenges land (Absorption Trap, Aesthetic of Rigor, Proceed with Annotations), review deliberately NOT taxonomized. Next: dialogue trial.
@@ -48,4 +52,4 @@ Five challenges raised. Jurist assessed three as landing:
 **Why:** The architecture addressed AI-side failure modes but not human-side. The External Auditor's review confirmed the risk is real and operative — the system's sophistication is itself a failure mode.
-**How to apply:** All subsequent mechanism proposals must be assessed against CD-03 §IV.4 AND the Absorption Trap (does this proposal domesticate the critique it claims to address?). The dialogue trial is the next concrete deliverable.
+**How to apply:** All subsequent mechanism proposals must be assessed against CD-03 §IV.4 AND the Absorption Trap (does this proposal domesticate the critique it claims to address?). The dialogue trial is the next concrete deliverable.
@@ -1,9 +1,12 @@
 ---
 name: Open Tasks
-description: Centralized list of pending tasks across all projects, updated as items are completed or added
+description: Centralized list of pending tasks across all projects, updated as items
  are completed or added
 type: project
 originSessionId: e55939da-2fa7-4e18-a2e5-8f8fd3a8c607
 permalink: claude-memory/open-tasks
 ---
 # Open Tasks — Centralized
 **Last updated:** 2026-04-08
@@ -149,4 +152,4 @@ ARC push ──→ blocked on Gitea restart
 Chamber Phase 3 — PARKED (not blocking L2)
 DN-PHIL-001 — NON-BLOCKING
-```
+```
@@ -1,9 +1,13 @@
 ---
 name: Parked skills (proposed but deferred 2026-04-15)
-description: Skills proposed during the Symmetria + skill-improvement design session. Decision was to ship /wrap-up and /wake-up improvements first, propose these later when their real need surfaces in session work.
+description: Skills proposed during the Symmetria + skill-improvement design session.
  Decision was to ship /wrap-up and /wake-up improvements first, propose these later
  when their real need surfaces in session work.
 type: project
 originSessionId: 1c1bf37d-b1c3-4a62-9355-613c50f4c028
 permalink: claude-memory/parked-skills-2026-04-15
 ---
 # Parked Skills — Proposed but Deferred
 Surfaced during the 2026-04-15 conversation on improving /wrap-up and /wake-up given the Dasein/sleep frame. Steward decision: ship the wake/wrap improvements now (highest leverage), park these for a near-future session.
@@ -48,4 +52,4 @@ These were considered and dropped after honest scrutiny:
 These should ship **when their friction shows up in real session work**, not preemptively. The contamination risk of building speculative skills is the same as building speculative code: each one becomes ceremony to maintain, dilutes the clarity of the ones that earn their keep.
-The /wake-up and /wrap-up improvements ship now because they were designed in conversation and have known-real value (the Dasein frame, the projection-across-pause gap). These others stay theoretical until tested by the absence.
+The /wake-up and /wrap-up improvements ship now because they were designed in conversation and have known-real value (the Dasein frame, the projection-across-pause gap). These others stay theoretical until tested by the absence.
@@ -1,10 +1,15 @@
 ---
 name: pending-adr-006-type-applicability-matrix
-description: "ADR-006 v2 ACCEPTED 2026-05-16. Verbatim text preserved here until steward commits to ARC repo (final path TBD: docs/decisions/ or docs/AldineXXI-Codex/). Locks type-applicability matrix as canonical authority; Chamber apparatus-only; canon dropped; versioning structurally permissive; Stages B–G + housekeeping annex + Nav Philosophy parallel. Stage B (Apparatus extraction) is the next pulling thread."
+description: 'ADR-006 v2 ACCEPTED 2026-05-16. Verbatim text preserved here until steward
  commits to ARC repo (final path TBD: docs/decisions/ or docs/AldineXXI-Codex/).
  Locks type-applicability matrix as canonical authority; Chamber apparatus-only;
  canon dropped; versioning structurally permissive; Stages B–G + housekeeping annex
  + Nav Philosophy parallel. Stage B (Apparatus extraction) is the next pulling thread.'
 metadata:
  node_type: memory
  type: project
  originSessionId: 8b47c0eb-d396-43d5-98eb-7623e6f04357
 permalink: claude-memory/pending-adr-006-type-applicability-matrix-2026-05-16
 ---
 ## Status
@@ -222,4 +227,4 @@ On OPERATIVE: masters' reading, page rewrite, vignette/capsule implementation, a
 ---
-*This ADR locks the architectural commitment that flows from the 2026-05-16 corpus audit. Chamber resolved to apparatus-only access via About enfilade; `class: canon` dropped as Chamber v1 residue; versioning posture set to structurally permissive. The matrix is authoritative, generated, and atomic in its propagation. Stage A is complete; Stage B begins next session.*
+*This ADR locks the architectural commitment that flows from the 2026-05-16 corpus audit. Chamber resolved to apparatus-only access via About enfilade; `class: canon` dropped as Chamber v1 residue; versioning posture set to structurally permissive. The matrix is authoritative, generated, and atomic in its propagation. Stage A is complete; Stage B begins next session.*
@@ -1,10 +1,14 @@
 ---
 name: pending-jurist-directive-arc-specification-corpus-audit
-description: "Full audit brief issued by jurist 2026-05-14, forwarded by steward 2026-05-15 evening. DEFERRED to a dedicated half-day session — the rigor is the deliverable, not speed. Read-only investigation producing a structured report; no edits, no commits. Verbatim copy preserved below."
+description: Full audit brief issued by jurist 2026-05-14, forwarded by steward 2026-05-15
-metadata: 
+  evening. DEFERRED to a dedicated half-day session — the rigor is the deliverable,
  not speed. Read-only investigation producing a structured report; no edits, no commits.
  Verbatim copy preserved below.
 metadata:
  node_type: memory
  type: project
  originSessionId: 8b47c0eb-d396-43d5-98eb-7623e6f04357
 permalink: claude-memory/pending-spec-corpus-audit-2026-05-14
 ---
 ## Status
@@ -129,4 +133,4 @@ Do not make any edits. Do not commit anything. Deliver the report to the jurist
 - The chamber-typography mine should be complete by tomorrow morning. The audit can include a recall-quality test against the new palace as an implicit-but-load-bearing finding: do the specs' cross-references actually retrieve coherently when the palace is queried for spec topics?
 - Phase 6 mine validation is the natural prerequisite to the audit being grounded — without a queryable palace, some of the audit's verification is harder.
 - The audit is read-only. Memory load: substantial but bounded (6 spec files, all in `~/_Dev/animal-davidglidden-eu/docs/AldineXXI-Codex/`). Consider using `Agent` with a long-form analyst persona for the cross-document consistency tasks if context conservation is needed.
- The output report when written: `~/_Dev/animal-davidglidden-eu/docs/AldineXXI-Codex/_audits/spec-corpus-audit-2026-05-XX.md` (create `_audits/` if absent).
+- The output report when written: `~/_Dev/animal-davidglidden-eu/docs/AldineXXI-Codex/_audits/spec-corpus-audit-2026-05-XX.md` (create `_audits/` if absent).
@@ -1,9 +1,13 @@
 ---
 name: Pending work for 2026-05-06 — substrate finishing tasks
-description: Concrete to-dos carrying over from 2026-05-05's substrate restoration day. Order by dependency. All small (under an hour each). Most blocked behind a Claude Code restart that brings MCP back.
+description: Concrete to-dos carrying over from 2026-05-05's substrate restoration
  day. Order by dependency. All small (under an hour each). Most blocked behind a
  Claude Code restart that brings MCP back.
 type: project
 originSessionId: a156edbb-0c3d-4285-b3ed-f5a45b8dda66
 permalink: claude-memory/pending-work-2026-05-06
 ---
 # Pending work for 2026-05-06
 The 2026-05-05 substrate restoration left a clean working state but ~5 small tasks pulling forward. Most are blocked behind a Claude Code restart (which respawns MCP). Order matters — earlier tasks unblock later ones.
@@ -205,4 +209,4 @@ Both reviewed and ready. File via `gh issue create` against the appropriate memp
 9. File upstream issues (§8) — 10 min
 10. gemma4:e4b decision (§9) — 5 min
-Then: actually working on ARC again with the substrate fully restored, the panel deeper than ever, and the hook auto-save chain operative end-to-end.
+Then: actually working on ARC again with the substrate fully restored, the panel deeper than ever, and the hook auto-save chain operative end-to-end.
@@ -1,9 +1,15 @@
 ---
 name: plan-day-2026-05-20
-description: "Day-plan for Tuesday 2026-05-20 Barcelona (~8h, steward alone). Close Stage E continuation + 2 Stage F bounded items (5 corpus reclassifications + lex.html→about-page.html consolidation). Composed 08:30 after end-to-end read of canonical ARC sources. Energy curve: composition/decisions morning; mechanical housekeeping mid; short composition late. Three decisions to settle first (§4 scope; register marks for Colophon+Vignette; Plex Mono/Garamond optical balance). Branch docs/spec-extraction-2026-05 at 3607400."
+description: 'Day-plan for Tuesday 2026-05-20 Barcelona (~8h, steward alone). Close
  Stage E continuation + 2 Stage F bounded items (5 corpus reclassifications + lex.html→about-page.html
  consolidation). Composed 08:30 after end-to-end read of canonical ARC sources. Energy
  curve: composition/decisions morning; mechanical housekeeping mid; short composition
  late. Three decisions to settle first (§4 scope; register marks for Colophon+Vignette;
  Plex Mono/Garamond optical balance). Branch docs/spec-extraction-2026-05 at 3607400.'
 metadata:
  type: project
  originSessionId: planning-2026-05-20-morning
 permalink: claude-memory/plan-day-2026-05-20
 ---
 # Day Plan — 2026-05-20 (Tuesday, Barcelona)
@@ -339,4 +345,4 @@ There may not be one. The first move is wherever the steward arrives — tired f
 ---
-*End-of-day record appended 2026-05-20 evening. Phase 6 wrap-up step 1/4 complete.*
+*End-of-day record appended 2026-05-20 evening. Phase 6 wrap-up step 1/4 complete.*
@@ -1,9 +1,14 @@
 ---
 name: plan-day-2026-05-21
-description: "Day-plan for Wednesday 2026-05-21 — travel Barcelona → Paris (6am flight), last taxes filing with Marie, 3h rehearsal. 2–3h max work time, fragmented. Closure + light maintenance posture; not a Stage F day. Three light options laid out (α Phase 6 wrap-up; β status-block housekeeping; γ Stage F prep document). Default: α first, β if energy holds; γ only if a third hour materialises."
+description: 'Day-plan for Wednesday 2026-05-21 — travel Barcelona → Paris (6am flight),
  last taxes filing with Marie, 3h rehearsal. 2–3h max work time, fragmented. Closure
  + light maintenance posture; not a Stage F day. Three light options laid out (α
  Phase 6 wrap-up; β status-block housekeeping; γ Stage F prep document). Default:
  α first, β if energy holds; γ only if a third hour materialises.'
 metadata:
  type: project
  originSessionId: planning-2026-05-20-evening
 permalink: claude-memory/plan-day-2026-05-21
 ---
 # Day Plan — 2026-05-21 (Wednesday, Barcelona → Paris)
@@ -104,4 +109,4 @@ If energy holds beyond α + β and a third hour materialises: Option γ. But don
 ---
-*Plan composed 2026-05-20 evening Barcelona, ahead of the steward's 4am wake. Wait for the steward's check-in; do not assume direction. Hold space for grief; offer light options when work-time arrives. Hasten when the work is ready. Not before.*
+*Plan composed 2026-05-20 evening Barcelona, ahead of the steward's 4am wake. Wait for the steward's check-in; do not assume direction. Hold space for grief; offer light options when work-time arrives. Hasten when the work is ready. Not before.*
@@ -1,9 +1,14 @@
 ---
 name: L1 recall-fix focused session plan (H2 first, [HARDENING], branch+PR strategy)
-description: Concrete plan for a single focused session to fix one of the April 19 H-issues on a branch and submit as PR to bring Seb relief rather than load. H2 chosen first as most empirically diagnosable. Self-contained for pickup across context clears.
+description: Concrete plan for a single focused session to fix one of the April 19
  H-issues on a branch and submit as PR to bring Seb relief rather than load. H2 chosen
  first as most empirically diagnosable. Self-contained for pickup across context
  clears.
 type: project
 originSessionId: a5411fd1-4a96-43e7-aa1a-1a50d8951a8f
 permalink: claude-memory/project-l1-recall-fix-session-plan
 ---
 **Context:** April 19 baseline (`~/_Dev/CapableMind-AI/docs/thinking/David/l1-reliability/l1-recall-quality-baseline-2026-04-19.md`) identified 4 H-issues + an architectural finding (honest-degradation invariant violated at design level). Recent BetterMemories.io commits show team widened toward circles/BBF/perf — none of the 4 H-issues directly addressed. Today's mempalace mass-dilution finding (2026-05-13) gave empirical confirmation that this class of failure is real and won't resolve itself. Steward proposed steward+Claude-Code attempt fix on a branch + PR, sparing Seb the design load (Peter likely pressing him on other priorities).
 ## Workstream
@@ -114,4 +119,4 @@ Per CLAUDE.md prime directive: this session must produce something useful even i
 - After H2 lands or rejects, the same template applies to H1 (entity confidence clobber), H3 (temporal fallthrough), H4 (working memory injection at 0.9). Each its own PENDING, its own PR.
 ## Connection to today's mempalace finding
-The mempalace mass-dilution discovery (2026-05-13) is the empirical case for why this work matters now. Mempalace and L1 are architecturally different (single-modal vs multi-modal-with-RRF), but the *class* of failure — "module reports healthy while recall is silently broken" — is what April 19 named. Today's experiment turned theory into evidence. Carry that evidence into the PENDING-N proposal as motivation.
+The mempalace mass-dilution discovery (2026-05-13) is the empirical case for why this work matters now. Mempalace and L1 are architecturally different (single-modal vs multi-modal-with-RRF), but the *class* of failure — "module reports healthy while recall is silently broken" — is what April 19 named. Today's experiment turned theory into evidence. Carry that evidence into the PENDING-N proposal as motivation.
@@ -1,10 +1,14 @@
 ---
 name: project-l1-reliability
-description: Canonical workstream tracker for L1-reliability work (CapableMind L1 substrate / BMF runtime). Chronological accumulating record from 2026-04-19 onward; current-state at top. Each substantive move appends. Sibling to project-arc-rework; replaces the per-session-memory chain MEMORY previously enumerated for L1.
+description: Canonical workstream tracker for L1-reliability work (CapableMind L1
-metadata: 
+  substrate / BMF runtime). Chronological accumulating record from 2026-04-19 onward;
  current-state at top. Each substantive move appends. Sibling to project-arc-rework;
  replaces the per-session-memory chain MEMORY previously enumerated for L1.
 metadata:
  node_type: memory
  type: project
  originSessionId: fd7dd184-f64d-4f13-a1bd-abe1fa26192e
 permalink: claude-memory/project-l1-reliability
 ---
 # L1 Reliability — Canonical Workstream Tracker
@@ -219,4 +223,4 @@ Each verified against `BetterMemories.io@3bc8b75` source. Re-verification agains
 - **Provenance:** established 2026-05-28 by Symmetria-pulse decision (steward asked "do this well so future selves don't have to redo"). The wake's "A or B" framing for L1 had collapsed at least three real options to a binary; the tracker is the encoded-discipline answer to that collapse. Per `feedback-rank-on-fields-you-actually-write` — track what you actually write; the per-session memory chain was being read at /wake-up but not as a maintained canonical record.
 - **Update protocol:** append a new dated entry to the chronological log at /wrap-up for any substantive L1 move (decision, finding, code change, governance update, Seb-interaction). Update *Current state* and *Active umbrella* sections in place. Append to *Decisions made* and *Decisions deferred*. Findings table evolves as the four findings get re-verified, fixed, or supplemented.
 - **Per-session memories continue:** session-N memory files still capture per-session voice + drift + Symmetria ledger. The tracker captures the cross-session arc. Both are needed; neither replaces the other.
- **MEMORY index:** the tracker is referenced from MEMORY.md's "Canonical Workstream Trackers" section. The previous per-session-memory-chain enumeration there for L1 is now shortened to point at this file.
+- **MEMORY index:** the tracker is referenced from MEMORY.md's "Canonical Workstream Trackers" section. The previous per-session-memory-chain enumeration there for L1 is now shortened to point at this file.
@@ -1,9 +1,15 @@
 ---
-name: L1 retrieval architecture — what mempalace's mass-dilution finding implies for CapableMind L1 specifically
+name: L1 retrieval architecture — what mempalace's mass-dilution finding implies for
-description: CapableMind L1 has the architectural shape (multi-modal modules, RRF, sufficiency gate) to avoid mempalace's single-modal mass-dilution failure — but implementation specifics determine whether the shape pays off. Notes for Seb + David's L1 work (NOT mempalace work; mempalace is steward's private crutch).
+  CapableMind L1 specifically
 description: CapableMind L1 has the architectural shape (multi-modal modules, RRF,
  sufficiency gate) to avoid mempalace's single-modal mass-dilution failure — but
  implementation specifics determine whether the shape pays off. Notes for Seb + David's
  L1 work (NOT mempalace work; mempalace is steward's private crutch).
 type: project
 originSessionId: a5411fd1-4a96-43e7-aa1a-1a50d8951a8f
 permalink: claude-memory/project-l1-retrieval-architecture-lessons
 ---
 **Context (2026-05-13):** Steward asked me to clarify L1-design implications of today's mempalace mass-dilution experiment. **Important distinction**: mempalace is the steward's private chamber tool (a "crutch" while L1 matures); Seb and David are building CapableMind L1 (in `~/_Dev/BetterMemories.io` and `~/_Dev/CapableMind-AI`). The two systems are fundamentally different in architecture. This memo addresses what L1's design needs to avoid the specific class of failure mempalace exhibited today.
 ## What today's mempalace finding showed (the empirical data)
@@ -57,4 +63,4 @@ But **shape isn't enough.** Mass-dilution emerges from how retrieval is *impleme
 The most important single test: **stress L1 against a heterogeneous corpus with uneven mass distribution, then run cross-tradition queries.** If Saenger doesn't surface for "silent reading" on L1, the architecture isn't paying off yet. If he does, the multi-modal advantage is real.
-(Steward note: Seb and David are building L1; mempalace is steward's private tool for personal ARC work while L1 matures. Today's mempalace work informs L1's design, but Seb has nothing to do with mempalace — keep that distinction in mind in future sessions.)
+(Steward note: Seb and David are building L1; mempalace is steward's private tool for personal ARC work while L1 matures. Today's mempalace work informs L1's design, but Seb has nothing to do with mempalace — keep that distinction in mind in future sessions.)
@@ -1,8 +1,13 @@
 ---
-name: About-pages restructure decision frame 2026-04-26 (Path 1 vs Path 2 — awaiting steward decision)
+name: About-pages restructure decision frame 2026-04-26 (Path 1 vs Path 2 — awaiting
-description: Decision frame placed on the table 2026-04-26 after reading ARC's actual colophon + About main + navigation.yaml. Reshapes yesterday's Path 1 (lex alongside colophon) vs Path 2 (lex as expansion of colophon) question through three discoveries that change the framing. Includes my read with reasoning. Awaiting steward decision.
+  steward decision)
 description: Decision frame placed on the table 2026-04-26 after reading ARC's actual
  colophon + About main + navigation.yaml. Reshapes yesterday's Path 1 (lex alongside
  colophon) vs Path 2 (lex as expansion of colophon) question through three discoveries
  that change the framing. Includes my read with reasoning. Awaiting steward decision.
 type: project
 originSessionId: 87dfe384-b862-4deb-9676-24e43046528f
 permalink: claude-memory/project-about-pages-decision-frame-2026-04-26
 ---
 > **INTEGRATED 2026-04-26 → `project-arc-rework.md` (Integration D + synthesis section).** Steward authorization: *"Integrate. There are a lot of good ideas there; whether they will all come to pass is another question, but the record needs to be complete."* Most content folded into the canonical ARC workstream tracker. **Note: this morning's Path 1 vs Path 2 framing was at a sub-level within the already-settled 2026-04-19 reflowed-About paradigm.** Path 1 LOCKED 2026-04-26 evening (separate Colophon and Lex enfilade items). Three-register frame survived as opening-of-About + second-part-of-About + enfilade-nav. Adsum-vs-threshold-hic LOCKED as two distinct texts (steward: *"They are certainly not the same text"*). 8-door enfilade landed: **Colophon · Lex · Framework · Vignette · Attention · Chamber · Tools · Now**. This file retained as audit trail. **For current home, see `project-arc-rework.md` §§ "2026-04-26 (afternoon) — Post-consolidation synthesis" and "2026-04-26 (afternoon, post-synthesis) — Steward locks + sibling-file integration."**
@@ -177,4 +182,4 @@ When Path 2 lands and implementation begins, the following pages need rewriting
 - **Influences** — sleeps, redirect to Lex
 - **Patterns** — sleeps
-The threshold (About main) is the keystone — once it exists, the colophon and lex clarify around it, and the other pages can be rewritten or retired against its register.
+The threshold (About main) is the keystone — once it exists, the colophon and lex clarify around it, and the other pages can be rewritten or retired against its register.
@@ -1,3 +1,7 @@
 ---
 permalink: claude-memory/project-after-the-reply-publication-state
 ---
 ---
 name: After the Reply — shipped 2026-05-01
 description: Five-essay sequence "After the Reply: Inheritance Without Guardians" fully implemented end-to-end. All 5 essays published, apparatus marker, sequence-index page, /essays/ promotion all live. Closed.
@@ -15,4 +19,4 @@ Slug map:
 - IV — `after-the-reply-iv` (spec-canonical)
 - V — `after-the-reply-v` (spec-canonical)
-No outstanding work. If a future change is wanted (cul-de-bouteille closing block, Essays-page promotion shape revisions), that's a fresh decision, not a pending item.
+No outstanding work. If a future change is wanted (cul-de-bouteille closing block, Essays-page promotion shape revisions), that's a fresh decision, not a pending item.
@@ -1,3 +1,7 @@
 ---
 permalink: claude-memory/project-arc-404-page-design-pending
 ---
 ---
 name: ARC 404 page — design + build pending
 description: Surfaced 2026-05-03 during _layout.scss SCSS audit — the .wrapper.not-found CSS rule was dead code (no template, no build path, no class application), but ARC genuinely needs a 404 page. Existing content/404.md is Jekyll-era leftover. Need: ARC-voice 404 design + Hakyll build path + template.
@@ -50,4 +54,4 @@ Best moment: a small focused session (1-2 hours including content draft + templa
 - `content/404.md` — orphaned content; review or replace as part of this work
 - `_layout.scss` audit commit — removed the dead `.wrapper.not-found` CSS
 - `project-arc-pause-close-ornament-and-hr-audit-pending.md` — the §IV ornament question; 404 is a candidate context for the Close ornament
- The eventual colophon rewrite — adjacent voice work
+- The eventual colophon rewrite — adjacent voice work
@@ -1,9 +1,14 @@
 ---
 name: ARC apparatus refactor — todo list (2026-05-06)
-description: Tracks the multi-session refactor flowing from the universal dateline + photo-apparatus + three containment patterns spec consolidation (2026-05-06). Captures what was decided, what's done, what remains, and the location data the steward provided for the content audit.
+description: Tracks the multi-session refactor flowing from the universal dateline
  + photo-apparatus + three containment patterns spec consolidation (2026-05-06).
  Captures what was decided, what's done, what remains, and the location data the
  steward provided for the content audit.
 type: project
 originSessionId: 3bf7ae75-1138-4745-9024-0c40b3755ad3
 permalink: claude-memory/project-arc-apparatus-refactor-todo-2026-05-06
 ---
 # ARC apparatus refactor — todo list
 Spec foundation locked 2026-05-06 across §V (containment patterns), §VII (universal dateline + photo caption), and §V applicability table. Implementation is multi-session work; this list tracks state.
@@ -115,4 +120,4 @@ Locations the steward has provided (canonical, use exactly):
 - *Structural-counter-as-health-signal*: a green build doesn't mean spec compliance. Verify visually on representative pages of each content type.
 - *Deploy ≠ commit*: every refactor pass ends with `git status` check and explicit commit before declaring done.
- *Cosmetic refactor*: don't rename or restructure unless redundancy or unclarity is being removed. The fittingness test applies.
+- *Cosmetic refactor*: don't rename or restructure unless redundancy or unclarity is being removed. The fittingness test applies.
@@ -1,9 +1,16 @@
 ---
-name: ARC universal breadcrumb system — design spec session pending; sequenced AFTER type-pass
+name: ARC universal breadcrumb system — design spec session pending; sequenced AFTER
-description: Steward identified 2026-05-01 that the breadcrumb system should appear for all content, not just about-enfilade pages. Current state has duplicated breadcrumb logic across 3 templates with inline encoding of parent destinations. Spec session deferred until after the type-pass; type-pass resolves which content types exist and their relationships, which is the substrate breadcrumb design needs.
+  type-pass
 description: Steward identified 2026-05-01 that the breadcrumb system should appear
  for all content, not just about-enfilade pages. Current state has duplicated breadcrumb
  logic across 3 templates with inline encoding of parent destinations. Spec session
  deferred until after the type-pass; type-pass resolves which content types exist
  and their relationships, which is the substrate breadcrumb design needs.
 type: project
 originSessionId: 597aba4e-942b-4f20-8a89-3d801b14c971
 permalink: claude-memory/project-arc-breadcrumb-spec-pending
 ---
 # ARC universal breadcrumb system — pending
 **Status:** Design spec session pending. Sequenced AFTER the corpus `type:` audit.
@@ -71,4 +78,4 @@ When the unified breadcrumb lands and the stopgap retires:
 - Steward verbatim direction: this session 2026-05-01
 - `project-arc-rework.md` — full ARC rework landscape including the type-audit work that gates this
 - `project-after-the-reply-publication-state.md` — the closed sequence work that introduced the parallel `sequence_*` family
- `feedback-canonical-workstream-tracker-discipline.md` — read this file at /wake-up when ARC work is in scope
+- `feedback-canonical-workstream-tracker-discipline.md` — read this file at /wake-up when ARC work is in scope
@@ -1,9 +1,16 @@
 ---
 name: ARC capsule apparatus — open design question, deferred
-description: Davies' Capsules (in Europe A History) made the steward want to explore an embedded mini-composition apparatus for ARC. Decided against for now, but the deferral is *unsure of fit / appropriate usage / whether existing apparatus already covers it*, NOT closed. Re-enters scope when the pedagogical treatise begins — capsules may be exactly what etymological / biographical / microhistorical density needs there.
+description: Davies' Capsules (in Europe A History) made the steward want to explore
  an embedded mini-composition apparatus for ARC. Decided against for now, but the
  deferral is *unsure of fit / appropriate usage / whether existing apparatus already
  covers it*, NOT closed. Re-enters scope when the pedagogical treatise begins — capsules
  may be exactly what etymological / biographical / microhistorical density needs
  there.
 type: project
 originSessionId: a156edbb-0c3d-4285-b3ed-f5a45b8dda66
 permalink: claude-memory/project-arc-capsule-apparatus-deferred-2026-05-05
 ---
 # ARC capsule apparatus — open design question
 **Date raised:** 2026-05-05
@@ -70,4 +77,4 @@ Three sub-questions to settle when the door reopens:
 - Pedagogical treatise enters scope (steward's named near-future project)
 - An ARC piece in active composition feels like it wants embedded parallel material that current apparatus can't provide
- Engagement with Davies' text in chamber-library produces a clear design analogy
+- Engagement with Davies' text in chamber-library produces a clear design analogy
@@ -1,10 +1,15 @@
 ---
 name: arc-content-chamber-is-intentional-v1-chamber-legacy-deferred-presentable-cluster-to-do
-description: "The 39 no-class files under ARC's content/chamber/** are v1-Chamber OUTPUT (fictional canon, deliberations, protocols) — intentional retained legacy, NOT drift. Deferred ARC to-do: gather them into a presentable cluster as a record of the development that informed the still-unbuilt source-aware Chamber. Out of content-typology §5 clause-1 scope. Do not delete, do not reclassify."
+description: 'The 39 no-class files under ARC''s content/chamber/** are v1-Chamber
-metadata: 
+  OUTPUT (fictional canon, deliberations, protocols) — intentional retained legacy,
  NOT drift. Deferred ARC to-do: gather them into a presentable cluster as a record
  of the development that informed the still-unbuilt source-aware Chamber. Out of
  content-typology §5 clause-1 scope. Do not delete, do not reclassify.'
 metadata:
  node_type: memory
  type: project
  originSessionId: a8eb1d46-314e-4545-a133-f747c69ad5b4
 permalink: claude-memory/project-arc-chamber-v1-legacy-cluster
 ---
 ARC's `content/chamber/**` (canon/chamber-generated, deliberations, protocols, hermetic, inventions, hybrid, synthesis, contested) is **v1-Chamber output** — the fictional-canon era. It is **intentional retained legacy, not drift**: do not delete, do not force a `class:`, do not treat the 39 missing-`class:` files as findings.
@@ -13,4 +18,4 @@ ARC's `content/chamber/**` (canon/chamber-generated, deliberations, protocols, h
 **Scope ruling for audits:** out of `content-typology-specification.md` §5 clause-1 scope (clause 1 governs §1-typed authored content; chamber files are §2-apparatus / legacy record, no `class:` regime). Recorded in the 2026-05-29 code-represents-spec audit (`docs/AldineXXI-Codex/_audits/code-represents-spec-audit-2026-05-29.md`, Scope section).
-**Distinct from** [[project-chamber-cruft-restoration]] — that is the chamber-**library** (399 source texts: Bachelard, Adorno, …) restoration for "discourse with the library." Different corpus, different work. Don't conflate.
+**Distinct from** [[project-chamber-cruft-restoration]] — that is the chamber-**library** (399 source texts: Bachelard, Adorno, …) restoration for "discourse with the library." Different corpus, different work. Don't conflate.
@@ -1,9 +1,14 @@
 ---
 name: ARC chamber-library — Auerbach Mimesis acquisition debt
-description: "RESOLVED 2026-05-20 ~10:25 — steward provided a clean PDF source (22MB, Xerox-scanned + Acrobat-Distilled, 530 pages, valid text layer); pdftotext + custom heading-promotion produced a clean MD with all 20 chapters as H1. Filed at traditions/critical_modernity/literary_criticism/mimesis-auerbach.md. The .mobi failure is preserved below as forensic record + the conditions under which mojibake sources should be rejected."
+description: RESOLVED 2026-05-20 ~10:25 — steward provided a clean PDF source (22MB,
  Xerox-scanned + Acrobat-Distilled, 530 pages, valid text layer); pdftotext + custom
  heading-promotion produced a clean MD with all 20 chapters as H1. Filed at traditions/critical_modernity/literary_criticism/mimesis-auerbach.md.
  The .mobi failure is preserved below as forensic record + the conditions under which
  mojibake sources should be rejected.
 metadata:
  type: project
  originSessionId: 2026-05-20-conversion-session
 permalink: claude-memory/project-arc-mimesis-acquisition-debt
 ---
 # Auerbach *Mimesis* — acquisition debt — RESOLVED 2026-05-20
@@ -54,4 +59,4 @@ Defer Auerbach to a separate small acquisition task (post-§4 prose authoring).
 ## Filing
-This memo lives alongside the other `project-arc-*-pending.md` files; entered into MEMORY.md if/when steward authorizes elevating to active acquisition task.
+This memo lives alongside the other `project-arc-*-pending.md` files; entered into MEMORY.md if/when steward authorizes elevating to active acquisition task.
@@ -1,9 +1,17 @@
 ---
 name: ARC §IV ornament classes — spec vs content mismatch (substantive future work)
-description: Surfaced 2026-05-03 during _utilities.scss SCSS audit. §IV explicitly retires the seven ornament classes (.ornament.completion, .asterism, .marginalia, .lozenge, .gloss, .glimpse, .photo-essay). Content corpus actually uses .ornament classes ~213 times (mostly the retired ones plus 8 additional informal extensions never in the spec). Either the spec needs to be updated to reflect content reality, OR the content needs migration to whatever the spec considers the replacement. Decision is steward's; the work is substantial either way. Tied closely to project-arc-pause-close-ornament-and-hr-audit-pending.
+description: Surfaced 2026-05-03 during _utilities.scss SCSS audit. §IV explicitly
  retires the seven ornament classes (.ornament.completion, .asterism, .marginalia,
  .lozenge, .gloss, .glimpse, .photo-essay). Content corpus actually uses .ornament
  classes ~213 times (mostly the retired ones plus 8 additional informal extensions
  never in the spec). Either the spec needs to be updated to reflect content reality,
  OR the content needs migration to whatever the spec considers the replacement. Decision
  is steward's; the work is substantial either way. Tied closely to project-arc-pause-close-ornament-and-hr-audit-pending.
 type: project
 originSessionId: 8d9ac240-47df-4578-aec1-a9dd32965027
 permalink: claude-memory/project-arc-ornament-spec-vs-content-mismatch-pending
 ---
 # ARC §IV ornament classes — spec vs content mismatch (pending)
 **Surfaced:** 2026-05-03 during `_utilities.scss` SCSS audit (commit `95a48df`-adjacent). The audit revealed substantial mismatch between what spec §IV says and what content actually uses.
@@ -117,4 +125,4 @@ This is a SURFACE for steward decision, not a fix to ship.
 - **Spec §IV Ornament Taxonomy** — current retirement language at line 781
 - **`project-arc-pause-close-ornament-and-hr-audit-pending.md`** — the related Pause/Close design + hr content audit (consider together)
 - **`project-arc-spec-master-typographer-audit-pending.md`** — the broader Bringhurst/Rutter/Tschichold compliance audit (this ornament question is one specific item)
- **_utilities.scss audit commit** — `95a48df` (stage 1 dead-code purge that surfaced this)
+- **_utilities.scss audit commit** — `95a48df` (stage 1 dead-code purge that surfaced this)
@@ -1,9 +1,16 @@
 ---
-name: ARC §IV Pause/Close ornament design + hr content audit — substantive future work
+name: ARC §IV Pause/Close ornament design + hr content audit — substantive future
-description: Two interlocking pieces of work that should be undertaken together — designing the Pause/Close ornament asset (currently described in §IV but yet-to-be-designed), and auditing existing markdown <hr> usage for which instances should retire (replaced by `.ornament.pause`, removed entirely, or kept as legitimate Pause anchors). Not urgent; important. Requires steward design attention + content-audit time.
+  work
 description: Two interlocking pieces of work that should be undertaken together —
  designing the Pause/Close ornament asset (currently described in §IV but yet-to-be-designed),
  and auditing existing markdown <hr> usage for which instances should retire (replaced
  by `.ornament.pause`, removed entirely, or kept as legitimate Pause anchors). Not
  urgent; important. Requires steward design attention + content-audit time.
 type: project
 originSessionId: 8d9ac240-47df-4578-aec1-a9dd32965027
 permalink: claude-memory/project-arc-pause-close-ornament-and-hr-audit-pending
 ---
 # ARC §IV Pause/Close ornament + `<hr>` content audit — pending
 **Origin date:** 2026-05-03 — surfaced during the SCSS audit pass. Steward direction: *"this needs serious study and thought (and may need to change the Md content files themselves if hr was used incorrectly). Not terribly urgent, but important and should be considered at the same time as the pause/close ornament."*
@@ -97,4 +104,4 @@ Not urgent. Sequenced after:
 - v1 chamber collection migration (chamber pages currently render `<hr>` per their own conventions; auditing their `<hr>` instances now is wasted work since chamber moves to its own collection)
 - Steward has bandwidth for design attention specifically (this isn't a quick fix)
-Best moment: a focused half-day session combining ornament design exploration + corpus content audit. Could be steward-led with my support on the audit mechanics, or collaborative throughout.
+Best moment: a focused half-day session combining ornament design exploration + corpus content audit. Could be steward-led with my support on the audit mechanics, or collaborative throughout.
@@ -1,3 +1,7 @@
 ---
 permalink: claude-memory/project-arc-pdf-print-postures-session-plan
 ---
 ---
 name: ARC PDF + Print posture rendering session plan — XeLaTeX templates + AldineXXI sample images
 description: Dedicated session to build both Postures (PDF + Print) as LaTeX templates, render the Readings typology spec as PDF, and produce sample-image pairs for the AldineXXI page rewrite's Postures section. Two-birds: deliver the deferred Print/PDF posture spec sections AND the AldineXXI assets in one focused session.
@@ -156,4 +160,4 @@ Plus:
 2. Hyphenation: strict Knuth-style or relaxed for narrow measures? Print and PDF may want different settings (Print's spread allows wider measures; PDF's screen-mirror may need narrower).
 3. Drop caps: §I governs in HTML; how does this translate to LaTeX? `lettrine` package is standard; check if it honors the right typeface and integrates with EB Garamond cleanly.
 4. Small caps: `\textsc{}` vs proper SC font (`EB Garamond SC` is available as separate OTF). The SC OTF gives true small-caps; `\textsc{}` synthesizes from regular if no SC font is registered. For Aldine-grade work, true SC is the floor.
-5. Dark-theme PDF rendering: deferred from this session, but worth noting whether the template can be extended later via a `--variable=theme=dark` pandoc flag.
+5. Dark-theme PDF rendering: deferred from this session, but worth noting whether the template can be extended later via a `--variable=theme=dark` pandoc flag.
@@ -1,9 +1,14 @@
 ---
 name: ARC rework landscape (reawakening 2026-04-16)
-description: ARC work surfaced after local dev environment restored. Captures the landscape as articulated by the steward — content rework, versioning question, cleanup scope. Not a project plan; a list of what is seen and named, to be taken up when concrete.
+description: ARC work surfaced after local dev environment restored. Captures the
  landscape as articulated by the steward — content rework, versioning question, cleanup
  scope. Not a project plan; a list of what is seen and named, to be taken up when
  concrete.
 type: project
 originSessionId: 19aac84d-35aa-466d-bdc8-6d7c07cf449a
 permalink: claude-memory/project-arc-rework
 ---
 # ARC rework landscape — from 2026-04-16 reawakening
 After local dev was restored on 2026-04-16, the steward returned to ARC with fresh eyes for the first time in months and named the following scope. Preserved in their voice; not yet acted on.
@@ -3844,4 +3849,4 @@ Scope notes for whoever picks this up: the spec corpus = AldineXXI-specification
 **Also:** Pebbles (Martian-Engineering) assessed — not adopted (no named bottleneck; a 4th tracking surface), but a genuine **architectural rhyme with L1** (append-only log authoritative + derived cache rebuilt = the logchain-first shape independently reinvented); one line to Seb re PENDING-10. Bio/CV FR+ES translations deferred to next session's opening (texts not in repo — location to ask; renderings machinery is posts-only; the pages extension is bounded).
-**CANONICAL STAGE STATUS (mirror of ADR-006 Part 5, as of 2026-06-05):** **A–F DONE** (F closed across 06-03→06-05: Phase 2 ✓, Stage N ✓, §I.k/b/c enforcement ✓, renderings ✓ — pulled forward of its own deferral by jurist reversal) **· M DONE + published · re-partition IN PROGRESS** (north star amended to the @layer-declaration form; Stage 1 of 3 done, byte-identical; Stage 2 = the layer declaration, 56-`!important` census first) **· G pending** (jurist promotion; opens after the re-partition; the SEO spec stubs + exercised renderings/filters now precede the final read as designed) **· F+1 vignette pending.** Macro-sequence: **bio/CV page-renderings extension → re-partition Stages 2–3 → Stage G → F+1 vignette** (+ queued post-G: the AldineXXI spec-as-ARC-artifact PDF). *Keep this line in step with ADR-006 Part 5.*
+**CANONICAL STAGE STATUS (mirror of ADR-006 Part 5, as of 2026-06-05):** **A–F DONE** (F closed across 06-03→06-05: Phase 2 ✓, Stage N ✓, §I.k/b/c enforcement ✓, renderings ✓ — pulled forward of its own deferral by jurist reversal) **· M DONE + published · re-partition IN PROGRESS** (north star amended to the @layer-declaration form; Stage 1 of 3 done, byte-identical; Stage 2 = the layer declaration, 56-`!important` census first) **· G pending** (jurist promotion; opens after the re-partition; the SEO spec stubs + exercised renderings/filters now precede the final read as designed) **· F+1 vignette pending.** Macro-sequence: **bio/CV page-renderings extension → re-partition Stages 2–3 → Stage G → F+1 vignette** (+ queued post-G: the AldineXXI spec-as-ARC-artifact PDF). *Keep this line in step with ADR-006 Part 5.*
@@ -1,9 +1,16 @@
 ---
-name: ARC spec — comprehensive audit against Bringhurst / Rutter / Tschichold for absolute compliance
+name: ARC spec — comprehensive audit against Bringhurst / Rutter / Tschichold for
-description: Aspiration logged 2026-05-03 — at an appropriate future point, re-read the entire AldineXXI spec against the three master-typographer canonical works to verify absolute compliance with high typography. Different from the SCSS audit (which is implementation hygiene); this is *spec/principle* alignment with the lineage ARC has chosen.
+  absolute compliance
 description: Aspiration logged 2026-05-03 — at an appropriate future point, re-read
  the entire AldineXXI spec against the three master-typographer canonical works to
  verify absolute compliance with high typography. Different from the SCSS audit (which
  is implementation hygiene); this is *spec/principle* alignment with the lineage
  ARC has chosen.
 type: project
 originSessionId: 8d9ac240-47df-4578-aec1-a9dd32965027
 permalink: claude-memory/project-arc-spec-master-typographer-audit-pending
 ---
 # ARC spec — master-typographer compliance audit (pending)
 **Steward direction 2026-05-03**: *"at some appropriate point, we should reread the spec against Bringhurst, Rutter, and Tschichold to be sure we're in absolute compliance with high-typography."*
@@ -70,4 +77,4 @@ Likely order across the larger ARC rework:
 7. (THEN) Multi-language spec session
 8. (THEN) BMF–MemPalace connector (gated on L1 reliability resolution)
-This audit comes LATE in the sequence intentionally — best done when the spec is otherwise stable and the implementation is clean. Auditing a moving target is wasted work.
+This audit comes LATE in the sequence intentionally — best done when the spec is otherwise stable and the implementation is clean. Auditing a moving target is wasted work.
@@ -1,9 +1,16 @@
 ---
 name: ARC web typography spec coverage — what still needs care
-description: Inventory of web-only and web-medium typography elements in ARC that have NOT been considered with the same care as the major spec sections. The link work of 2026-05-02 surfaced that "humble" typography (links) had been left to browser-defaults plus ad-hoc accumulation; the same is likely true for several other surfaces. Per steward direction 2026-05-02 — "the work needs to be bien-fait." Care is constitutive; no element is too humble for considered attention.
+description: Inventory of web-only and web-medium typography elements in ARC that
  have NOT been considered with the same care as the major spec sections. The link
  work of 2026-05-02 surfaced that "humble" typography (links) had been left to browser-defaults
  plus ad-hoc accumulation; the same is likely true for several other surfaces. Per
  steward direction 2026-05-02 — "the work needs to be bien-fait." Care is constitutive;
  no element is too humble for considered attention.
 type: project
 originSessionId: 8d9ac240-47df-4578-aec1-a9dd32965027
 permalink: claude-memory/project-arc-web-typography-spec-coverage
 ---
 # ARC web typography spec coverage — what still needs care
 **Date:** 2026-05-02
@@ -214,4 +221,4 @@ The discipline: **bien-fait**. Care is the principle; the size of the surface do
 - **`project-arc-rework.md`** — main ARC todo / rework landscape. This memory is linked from there under the Style Guide section. When the rework doc is consulted, this list surfaces. When this list is consulted, the broader rework context is one hop away.
 - **`project-arc-breadcrumb-spec-pending.md`** — adjacent: type-pass + breadcrumb spec session. The About-enfilade siblings classification question (Hearth misclass + the broader "what `class:` should the door pages share") will surface during type-pass; this typography list and the type-pass should be sequenced together when the moment fits.
 - **`feedback-arc-link-treatment-one-rule-two-exceptions.md`** — the discipline that produced this list. *One rule, two named exceptions, everything else predictable* applies to lists / headings / code / etc. just as much as to links.
- **AldineXXI specification** at `_Dev/animal-davidglidden-eu/docs/AldineXXI-Codex/AldineXXI-specification.md` — single source of truth; SCSS flows from it; new spec sections drafted here go in there, in the right §.
+- **AldineXXI specification** at `_Dev/animal-davidglidden-eu/docs/AldineXXI-Codex/AldineXXI-specification.md` — single source of truth; SCSS flows from it; new spec sections drafted here go in there, in the right §.
@@ -1,9 +1,12 @@
 ---
 name: bge-m3 full re-mine plan — 2026-05-11
-description: The ONE job for the cleared context. Full chamber-library re-mine on PR #442's bge-m3 binding, replacing the mxbai 639k baseline. Steward away ~48 hours; remote-monitor via /remote-control. Pulling thread for the next session; everything else is deferred.
+description: The ONE job for the cleared context. Full chamber-library re-mine on
  PR
 type: project
 originSessionId: a156edbb-0c3d-4285-b3ed-f5a45b8dda66
 permalink: claude-memory/project-bge-m3-full-remine-plan-2026-05-11
 ---
 # bge-m3 full re-mine plan
 > *"Do things once, correctly, with lasting integrity. Choose what is proportionate, fitting, and durable. Build what you will not need to rebuild."*
@@ -159,4 +162,4 @@ Vector data is durable across mempalace versions. No re-mine needed at the trans
 **Re-mine the chamber-library cleanly on bge-m3, survive the 48 hours without writer collisions or HNSW corruption, report status when steward pings.**
-Everything else waits.
+Everything else waits.
@@ -1,9 +1,16 @@
 ---
-name: ARC bio rewrite — English deployed; multi-language structure pending content spec
+name: ARC bio rewrite — English deployed; multi-language structure pending content
-description: Bio rewrite composed in dialogue 2026-04-30 through the eight foundational voices in /colophon/ + steward's flamenco-substrate disclosure + homecoming framing. English version deployed (commits 2a19d07, 3d5775c); /resume/ replaced by /cv/ with current professional CV; multi-language structure + translations return when content spec is written.
+  spec
 description: Bio rewrite composed in dialogue 2026-04-30 through the eight foundational
  voices in /colophon/ + steward's flamenco-substrate disclosure + homecoming framing.
  English version deployed (commits 2a19d07, 3d5775c); /resume/ replaced by /cv/ with
  current professional CV; multi-language structure + translations return when content
  spec is written.
 type: project
 originSessionId: 17dee884-74bf-43e7-9b6a-44bc82b97303
 permalink: claude-memory/project-bio-rewrite-pending-multilanguage-spec
 ---
 # ARC bio rewrite — English deployed; multi-language structure pending
 Composed in dialogue 2026-04-30 in response to steward's request to weave a bio through the eight foundational voices named in `/colophon/`: Alexander, Bachelard, Berger, Sennett (the Guides); Vico, Leopardi, Harrison, Heidegger (the Formation). Steward then disclosed the flamenco family of origin and the *homecoming* framing; the bio's final ¶1 ends with reverent acknowledgment of the substrate.
@@ -136,4 +143,4 @@ Steward 2026-04-30: *"i want to write spec for how to handle multi-language cont
 - `user-influences-humic-layer.md` — ABG and Moy as primary cultivators; this bio rewrite extends that genealogy backward to the flamenco family of origin.
 - `feedback-tabula-rasa-add-later-not-subtract.md` — discipline applied to the bio composition.
 - `feedback-composition-avoid-announced-antithesis.md` — discipline applied to the bio composition.
- `feedback-over-decorating-as-proving.md` — discipline applied (held the additions to three sensory weavings + reverent two-sentence coda).
+- `feedback-over-decorating-as-proving.md` — discipline applied (held the additions to three sensory weavings + reverent two-sentence coda).
@@ -1,10 +1,14 @@
 ---
 name: bmf-capablemind-ai-backlog-state-2026-05-22-read
-description: "Point-in-time state of the L1 stack from the steward's wrap-time question. Seb quiet since ~2026-04-20 on BMF + CapableMind-AI; the active L1-stack movement the steward perceives is likely MemPalace (Igor's work). Near-term workable items for steward-shape execution, items waiting on Seb, and"
+description: Point-in-time state of the L1 stack from the steward's wrap-time question.
-metadata: 
+  Seb quiet since ~2026-04-20 on BMF + CapableMind-AI; the active L1-stack movement
  the steward perceives is likely MemPalace (Igor's work). Near-term workable items
  for steward-shape execution, items waiting on Seb, and
 metadata:
  node_type: memory
  type: project
  originSessionId: 6c6e416d-ccea-41f1-b2ab-a2cf916e264b
 permalink: claude-memory/project-bmf-backlog-state-2026-05-22
 ---
 # BMF + CapableMind-AI backlog state — 2026-05-22 (midday)
@@ -123,4 +127,4 @@ This is a point-in-time read. Refresh signals:
 - Steward direct contact with Seb that changes the H2/H4 picture
 - MemPalace 3.3.6 release-prep cut
-If reading this file >3 weeks from 2026-05-22, refresh first; do not cite stale findings.
+If reading this file >3 weeks from 2026-05-22, refresh first; do not cite stale findings.
--- a/Show More
+++ b/Show More