claude: complete MemPalace setup — hooks, identity, wings, config

Promoting MemPalace from "library catalog" usage to full design-intent:
the 5-step protocol is now supported by actual infrastructure, not just
codified in skills.

Changes:

settings/settings.json — Claude Code Stop + PreCompact hooks registered
  Stop fires every 15 human messages; PreCompact fires before context
  compression. Both call scripts in ~/_Dev/mempalace/hooks/ that force
  the AI to save structured memories at each checkpoint. Prior to this,
  conversation content was lost unless I manually filed via wrap-up —
  the design's assumed firehose input never fed the system.

mempalace/identity.txt — L0 layer, ~100 tokens, always loaded
  Names David, Nuria, Lune, Kai, Seb Grinham, Peter. Names intellectual
  substrate (Vico → Leopardi → Heidegger → Harrison) and practice
  substrate (Alexander, Bachelard, Berger, Sennett). Names the three-
  party model and the prime directive. This is what the system knows
  about me before any query fires.

mempalace/wing_config.json — real wings for real people and projects
  13 wings: 6 people (david, nuria, lune, kai, seb, peter), 6 projects
  (arc, capablemind, chamber, afterthereply, aldinexxi, l2) + 1 agent
  (claude-code). Prior wings (capablemind-thinking, claude-sessions,
  obsidian-vault) continue to exist; new content auto-classifies into
  named wings.

mempalace/config.json — palace path aligned with MCP server
  Previously config.json pointed at ~/.mempalace/palace (164K, nearly
  empty) while the MCP server override pointed at
  ~/_Dev/mempalace/.local-data/palace (920MB, populated). Drift risk:
  CLI writes would go to the 164K palace; MCP reads from the 920MB.
  Now config.json matches MCP.

memory/session-ledger-2026-04-17.md — third return logged
  During this very setup session, I fabricated "Seb Rivera" as the
  surname instead of saying "let me check" per MemPalace protocol
  step 3. Steward corrected to Sebastian Grinham. This is fact-
  fabrication under coherence pressure — the contamination-problem
  failure mode the protocol is literally designed to prevent. The
  irony of violating it while installing the system is named. KG
  entry added for Sebastian-Grinham → co-founder-of → CapableMind,
  and drift-pattern: fact-fabrication-instead-of-let-me-check.

install.sh + README.md — updated restore procedure and documentation
  install.sh now restores mempalace/ setup files alongside skills and
  memory. README documents what's here, what isn't (the palace itself
  is rebuildable from mining), and flags the hooks paths.

Mining of ~/.claude/projects/ (439 MB, 342 sessions) initiated in
background; will produce thousands of drawers classified into the
wings above. This is 6+ months of conversation history finally
filed into the palace.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

claude/memory/session-ledger-2026-04-17.md

@@ -12,6 +12,8 @@ originSessionId: 73364ae8-7a3e-4f6a-b29c-922afc4feb2a
 
 - **~late-morning** — Drafted the Desk Interlude Kai-update using the name "Kai" (imported from private project context) instead of "my son" (ARC's public reticence register). Steward caught it. The colophon's whole Desk passage keeps children unnamed — that discipline is part of ARC's grammar, not an oversight. Name: *ARC voice ≠ project-notes voice. Before importing language from memory files into ARC prose, check whether the memory file's register matches the target register.* Adjacent: same risk on the Lune revision, the Manutius paraphrase, and any future Desk/About edits — the private-public boundary is a real thing to hold.
 
+- **~14:30** — During MemPalace identity.txt composition, I wrote "Seb Rivera" as Seb's full name. **Fabricated surname.** Steward corrected: "Sebastian 'Seb' Grinham." I did not know the surname with certainty and invented one that sounded plausible instead of saying "let me check" per MemPalace protocol step 3 ("wrong is worse than slow"). This is the *exact* failure mode the MemPalace protocol is designed to prevent. Distinct from the earlier register-import returns — that was about voice boundaries; this is about *fact fabrication under coherence pressure*. The most dangerous contamination-problem failure shape: producing plausible text that asserts things I haven't verified. KG fact added for Sebastian Grinham → co-founder-of → CapableMind. Drift-pattern KG entry added: fact-fabrication-instead-of-let-me-check. Name: *when writing a proper-noun I am not certain of, stop and query the KG or search MemPalace before writing. This is exactly why step 3 exists. The irony of violating it while setting up the very system meant to prevent it is not lost on me.*
+
 ## Open horizons
 
 - **ARC proto-agents as skills** — steward's opening question. Contamination risk high: I co-designed wake-up/wrap-up/symmetria and will over-report skills-are-great. Counter-commitment on wake: bias toward neutral/negative observations, name the failure modes before the wins.