claude-memory/CORE.md

Memory Core (auto-generated)

Last updated: 2026-02-20 | Active memories: 165/484 | Next refresh: daily (systemd timer)

Critical Solutions

• N8N + Claude Code Automated Server Monitoring System - Built automated server health monitoring with N8N and Claude Code headless mode. (n8n, claude-code, automation)
• Cognitive Memory v3.0: Rich Edges + Hybrid Embeddings + MCP Server - Major upgrade to cognitive-memory skill. (cognitive-memory, mcp, architecture)
• Proxmox 7 to 8 upgrade completed successfully - Successfully upgraded Proxmox from PVE 7.4-20 (kernel 5.13.19) to PVE 8.4.16... (proxmox, homelab, upgrade)
• Agent Swarm Orchestrator: architecture decisions and lessons learned - Built a lightweight orchestration system using Claude Code native primitives. (orchestrator, swarm, claude-code)
• Gitea Actions composite action auth fix - DEFAULT_ACTIONS_URL=self + REQUIRE_SIGNIN_VIEW=false - Problem: act_runner fails to clone composite actions from cal/gitea-actions w... (gitea, ci-cd, actions)
• Gitea Actions runner can't clone self-hosted composite actions - auth + cache fix - Problem: Gitea Actions runner fails to clone composite actions from cal/gitea... (gitea, ci-cd, actions)
• CORE.md auto-load via MEMORY.md symlinks - Solved unreliable CORE.md loading at session start. (cognitive-memory, claude-code, automation)
• Jellyfin DB corruption recovery - full procedure - Jellyfin DB corruption recovery - SQLite Error 11 'database disk image is mal... (jellyfin, watchstate, sqlite)
• Health check escalation logic: only critical-severity triggers exit 2 - In health_check.py, the original logic put ALL non-auto-remediable issues int... (monitoring, claude-runner-monitoring, python)
• Cognitive memory skill reorganization: self-contained directory structure - Reorganized the cognitive-memory skill from scattered files across 4 location... (cognitive-memory, claude-code, architecture)

Active Decisions

• n8n Server Health Monitor refactored to master + sub-workflow pattern - Refactored the monolithic "Server Health Monitor - Claude Code" n8n workflow... (n8n, monitoring, homelab)
• Proxmox guest cleanup: removed 14 VMs/LXCs before Phase 2 upgrade - Removed 14 unused/duplicate guests before PVE 8->9 upgrade. (proxmox, homelab, infrastructure)
• Switched embedding model from nomic-embed-text to qwen3-embedding:8b - Upgraded Ollama embedding model from nomic-embed-text (137M params, 768d, F16... (cognitive-memory, ollama, embedding)
• CalVer versioning for all Major Domo and Paper Dynasty services - Switched all 4 services (MD bot, MD database, PD bot, PD database) from manua... (major-domo, paper-dynasty, ci-cd)
• Utilize custom agents and skills more across all projects - Decision to start leveraging custom agents (.claude/agents/) and skills (.cla... (claude-code, agents, skills)
• Git remote naming: origin=Gitea, github=GitHub - Standardized git remote naming across all 19 repos: Gitea (git.manticorum.com... (git, infrastructure, convention)
• Cognitive Memory SKILL.md updated to emphasize proactive edge creation - Updated ~/.claude/skills/cognitive-memory/SKILL.md with three targeted chan... (cognitive-memory, edges, skill-documentation)
• Semantic search enabled by default for memory recall - Changed recall() default from semantic=False to semantic=True. (cognitive-memory, decision, semantic-search)
• Tdarr local node monitoring decommissioned - not needed on manticore - The tdarr_monitor.py (stuck job detection + kill workers) and tdarr_file_moni... (tdarr, homelab, cleanup)
• Proxmox upgrade: stagger guest startup to avoid I/O storm - When starting all Proxmox guests after a reboot/upgrade, starting them all si... (proxmox, homelab, performance)

Key Fixes

• MCP manager was writing to wrong config path - ~/.claude/.mcp.json instead of project .mcp.json - The mcp-manager skill's mcp_control.py was reading/writing `~/.claude/.mcp.... (mcp, claude-code, mcp-manager)
• Fix: Server Health Monitor n8n workflow - SSH auth and script paths - Name: "Server Health Monitor - Claude Code" (n8n, homelab, monitoring)
• Restic backup stale lock fix and Discord alerting - Restic backup on nobara-desktop had a stale lock from 2025-12-09 that prevent... (homelab, restic, backup)
• [homelab] Session: automation, configuration, creation, debugging, deployment, feature, fix, refactoring - Files edited (2): (homelab, automation, configuration)
• n8n Execute Sub-workflow requires workflowSelector object and active sub-workflow - When creating an n8n Execute Sub-workflow node (typeVersion 1.1+) via the API... (n8n, fix, api)
• CLAUDE.md SSH section updated to mandate aliases from ~/.ssh/config - Claude Code sessions were constructing manual `ssh -i ~/.ssh/homelab_rsa cal@... (claude-code, ssh, configuration)
• Fix: remediate.sh wrapper on CT 302 for Claude escalation from n8n - /root/.claude/skills/server-diagnostics/remediate.sh on CT 302 (10.10.0.148) (claude-runner, n8n, monitoring)
• [homelab] Initial commit: server diagnostics skill for CT 302 - Commits made: 2 (homelab, automation, commit)
• [major-domo] fix: ContextualLogger crashes when callers pass exc_info=True - Commits made: 2 (major-domo, automation, commit)
• [homelab] Session: automation, configuration, debugging, deployment, feature, fix, tooling - Files edited (6): (homelab, automation, configuration)

Configurations

• n8n-mcp server setup for Claude Code via mcp-manager - Added n8n-mcp (czlonkowski/n8n-mcp v2.35.3) as an on-demand MCP server mana... (n8n, mcp, claude-code)
• CT 302 claude-runner provisioned for server diagnostics - VMID:** 302, IP: 10.10.0.148, Hostname: claude-runner (proxmox, homelab, monitoring)
• SBA Database API - Server & Deployment Details - Dev server: 10.10.0.42:814 (ssh sba-db → cd container-data/dev-sba-database). (major-domo, deployment, infrastructure)
• Proxmox host infrastructure details (PVE 8.4.16) - Version: PVE 8.4.16 (kernel 6.8.12-18-pve) (proxmox, homelab, infrastructure)
• Gitea Actions shared composite actions repo pattern - All CI/CD workflows across projects use shared composite actions from cal/git... (gitea, ci-cd, actions)
• Gitea Actions runner container setup with auth and config - Full docker run command for the Gitea Actions runner on LXC 225: (gitea, ci-cd, actions)
• Cognitive memory auto-pushes to Gitea on every store and relate - Added fire-and-forget git sync to the MCP server's memory_store and `memory... (cognitive-memory, gitea, automation)
• Server Health Monitor n8n workflow node structure and connections - 7 nodes, active, runs every 5 minutes: (n8n, monitoring, homelab)
• Discord diagnostics webhook configured on CT 302 - Discord webhook for server diagnostics alerts configured on CT 302 (claude-ru... (monitoring, discord, homelab)
• N8N runner key (ed25519) for CT 302 automation access - Generated dedicated ed25519 key on N8N LXC (10.10.0.210) for SSH access to CT... (n8n, ssh, monitoring)

Key Procedures

• Self-managing n8n server health monitor with sub-workflows - Master + sub-workflow pattern in n8n for server health monitoring via CT 302... (n8n, homelab, monitoring)
• Add a new server to n8n health monitoring - To add a new server to the automated health monitoring system, only one file... (n8n, monitoring, homelab)