Properties
category: reference
tags: [claude-code, memory, process, backup]
last_updated: 2026-03-18
confidence: high

Claude Code Project Memory Backup

Backup of Claude Code's persistent memory for the otterwiki/robot.wtf project. These memories guide how Claude Code operates across sessions.

Architecture

Repos and Tools

All repos live under /Users/sderle/code/otterwiki/. The parent directory is NOT a git repo.

F/OSS Repos:

  • otterwiki/ — Fork of redimp/otterwiki (wiki engine). Upstream: github.com:redimp/otterwiki. Our remote: github.com:schuyler/otterwiki (SSH). Deploy branch: wikibot-io. Feature branches: feat/* off wikibot-io.
  • otterwiki-api/ — REST API plugin. Remote: github.com:schuyler/otterwiki-api. Branch: main.
  • otterwiki-semantic-search/ — FAISS + ONNX MiniLM semantic search plugin. Remote: github.com:schuyler/otterwiki-semantic-search. Branch: main.
  • otterwiki-mcp/ — MCP server plugin (12 tools). Remote: github.com:schuyler/otterwiki-mcp. Branch: main.

Private Repos:

  • robot.wtf/ — Platform app server, auth, MCP sidecar, API, Ansible deploy. Remote: github.com:schuyler/robot.wtf.
  • wikibot-io/ — AWS SaaS predecessor (Pulumi IaC, Lambda packaging). Remote: github.com:schuyler/wikibot-io. AWS profile: wikibot (us-east-1).
  • deploy/ — Legacy 3GW deploy config. No git remote.
  • dev-wiki/ — Local dev wiki git repo.

Deploy flow (robot.wtf): ansible-playbook ansible/deploy.yml -i ansible/inventory.yml from robot.wtf repo.

MCP Servers (DO NOT CONFUSE)

  • dev-wiki (mcp__dev-wiki__*) = dev.robot.wtf — platform dev instance
  • thirdgulfwar (mcp__thirdgulfwar__*) = 3gw.robot.wtf — research wiki (LXC/Caddy)
  • claude_ai_Third_Gulf_War_research_wiki — same as thirdgulfwar, via Claude.ai connector

MCP Authentication

  • MultiAuth composes InMemoryOAuthProvider (OAuth 2.1 for Claude.ai) + StaticTokenVerifier (bearer token for Claude Code)
  • OAuth is the transport layer — Caddy basic auth on /authorize* is the actual gate
  • Tokens are in-memory; server restart forces re-auth
  • Claude.ai URL: https://mcp.3gw.robot.wtf/mcp (must include /mcp!)
  • Claude.ai does NOT support bearer token auth — only OAuth

VPS Infrastructure

wiki-1 (app server): ssh sderle@192.168.77.107

  • Services: otterwiki (8000), MCP sidecar (8001), API (8002), auth (8003)
  • Data: /srv/data/robot.db, /srv/data/wikis/, /srv/data/robot.env
  • Ansible: /Users/sderle/code/otterwiki/robot.wtf/ansible/

proxy-1 (Caddy): ssh sderle@robot.wtf

  • Manages TLS for robot.wtf, *.robot.wtf
  • Read-only only. Managed by separate Ansible repo.

3GW (local network): LXC container at 3gw.lan, zmx session 3gw

Process Rules

Agent Delegation Model

  • Opus orchestrates, Sonnet implements. No Opus manager layer.
  • ALL subagents run in background (run_in_background: true), no exceptions.
  • Worktree isolation required for code changes. Parent dir is NOT a git repo — must scope to specific repo.

Implementation Workflow

See Design/Implementation_Workflow for full process. Pipeline: Plan (Opus) → Plan Review (Sonnet) → Implement (Sonnet, worktree) → Chico review → Zeppo review → Fix → Re-review → Verify.

Feedback Rules (Hard-Won Lessons)

Rule of Two: Never accept a subagent finding without independent verification. A test agent once reported a 404 bug using the wrong URL — it was accepted and written into the wiki before anyone checked.

Fix, don't defer: Never suggest carrying known debt when the fix is bounded and small. Default is always: fix it now.

Self-sufficient: Don't ask questions you can answer by reading code, wiki, or design docs. Only escalate when genuinely blocked or when a decision requires user judgment.

Reproducible VPS: NEVER mutate the VPS via SSH. All changes through Ansible. This has been corrected five times. Allowed via SSH: read-only diagnostics only (journalctl, systemctl status, curl, cat, ls, sqlite3 SELECT).

Pre-merge verification: Always dispatch a verification agent before merging any subagent's branch. Read the diff, confirm changes match spec, run tests.

TDD by default: Red/green TDD whenever possible. Write failing test first, then implementation. Discuss with user before skipping.

No ignored failures: If a test is failing before your changes, fix it or flag it. Don't normalize broken tests.

Plan Review: Always dispatch a Plan Review agent after Plan, before implementation. Include deployment context (gunicorn timeout, etc.).

Opus never implements: Always dispatch a subagent, even for trivial tasks. Opus holds the steering wheel, not the wrench.

Question the premise: When a finding says X is slow, ask whether X should exist at all before optimizing. The bcrypt linear scan fix was "add a prefix column" when the answer was "don't use bcrypt for API tokens."

Pre-flight checklist: Before dispatching worktree agents: (1) verify CWD is inside target git repo, (2) confirm plan-review loop complete, (3) confirm task dependencies wired.

Otterwiki config override: update_app_config() loads DB preferences and overwrites Flask config. DB wins over settings.cfg. Always seed values via _init_wiki_db().

Read memories first: Read MEMORY.md and Implementation_Workflow wiki page before dispatching any agents. Create full task list before launching first agent.

Project Context

Per-Wiki SQLite DB (COMPLETED)

Each wiki gets /srv/data/wikis/{slug}/wiki.db. Resolver swaps DB per-request. _init_wiki_db() seeds SITE_NAME, access levels, AUTH_METHOD, schema version.

Cold Start / Performance

Lambda cold starts: ~4.5s mean (512MB, Python 3.12). .pyc files make it WORSE (+700ms from extra decompression). CDN caching is the path forward, not Lambda optimization.

Shared Worktree Trial

Trial file-level parallelism: multiple agents writing different files in a shared worktree. Use EnterWorktree to create shared worktree, dispatch agents to same path with assigned files.

User Preferences

  • Concise, no sycophancy, no fake enthusiasm
  • Bias toward minimal solutions
  • Uses plan mode for non-trivial changes
  • Time is the most precious resource
  • git add -A is forbidden

Reference Pointers