f07022ca60
Implements automatic memory extraction when Claude Code sessions end: - Add SessionEnd hook to hooks.json with 120s timeout - Create session-end.sh wrapper that parses hook input and runs summarizer - Create summarize-transcript.py that: - Loads transcript from Claude's storage - Skips trivial sessions (<3 user messages) - Extracts paths/facts via heuristics - Uses Claude CLI (subscription auth) for decisions/preferences - Saves to memory files with deduplication - Updates history index with summarized flag Uses `claude -p --model haiku` for LLM extraction, leveraging existing subscription credentials instead of requiring API key. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Will's Claude Code Configuration
Personal assistant and multi-agent system for homelab management.
Quick Start
# Start a session
claude
# Use the personal assistant
/pa what's on my calendar today?
# Check system health
/sysadmin:health
# Check cluster status
/k8s:cluster-status
Features
| Feature | Description |
|---|---|
| Personal Assistant | Natural language interface via /pa |
| Gmail Integration | Read emails, check urgent, search |
| Calendar Integration | View agenda, check schedule |
| Kubernetes Management | Cluster health, deployments, diagnostics |
| System Administration | Health checks, updates, autonomy control |
| Usage Tracking | Session statistics and history |
Directory Structure
~/.claude/
├── .claude-plugin/ # Plugin manifest (for distribution)
├── agents/ # Agent persona definitions
├── automation/ # Systemd timers, scripts
├── commands/ # Slash commands (/pa, /gcal, etc.)
├── hooks/ # Event handlers (SessionStart, PreCompact)
├── mcp/ # MCP integrations (Gmail, delegation)
├── skills/ # Agent skills (gmail, gcal, k8s, sysadmin)
├── state/ # Shared state files (JSON)
├── workflows/ # Workflow definitions (design docs)
├── CLAUDE.md # Shared memory for all agents
└── README.md # This file
Each directory has its own README with details.
Key Commands
| Command | Description |
|---|---|
/pa <request> |
Natural language request to personal assistant |
/gcal [today|tomorrow|week] |
Calendar agenda |
/usage [today|week|month] |
Usage statistics |
/sysadmin:health |
System health check |
/sysadmin:update |
Package updates |
/k8s:cluster-status |
Kubernetes health |
/k8s:diagnose |
Troubleshoot K8s issues |
Skills (Auto-Invoked)
Skills are triggered automatically based on your request:
| Trigger | Skill |
|---|---|
| "check my email" | gmail |
| "what's on today" | gcal |
| "cluster status" | k8s-quick-status |
| "system health" | sysadmin-health |
| "usage stats" | usage |
Hooks
| Event | Action |
|---|---|
| Session Start | Load context, check pending items |
| Pre-Compact | Remind to preserve important context |
Agent Hierarchy
Personal Assistant (Opus)
└── Master Orchestrator (Opus)
├── linux-sysadmin (Sonnet)
├── k8s-orchestrator (Opus)
│ └── k8s-diagnostician, argocd-operator, etc.
└── programmer-orchestrator (Opus)
└── code-planner, code-implementer, code-reviewer
Configuration
| File | Purpose |
|---|---|
CLAUDE.md |
Shared instructions for all agents |
settings.json |
Claude Code settings |
state/component-registry.json |
Routing and discovery |
state/autonomy-levels.json |
Permission levels |
Plugin Installation
This configuration can be installed as a plugin:
# Add as local marketplace
/plugin marketplace add ~/.claude
# Install
/plugin install will-homelab@will-homelab-dev
Maintenance
# Validate component registry
python3 ~/.claude/automation/validate-registry.py
# Regenerate registry from files
python3 ~/.claude/automation/generate-registry.py
# Check systemd timers
systemctl --user list-timers
License
MIT