Stats
Actions
Tags
Help us improve
Share bugs, ideas, or general feedback.
From software-of-you
Runs health checks on SQLite database integrity, OAuth tokens, HTML views, backups, and sync freshness; auto-repairs issues and generates visual dashboard. Use /patrol for full patrol or /health for status.
npx claudepluginhub kmorebetter/better-software-of-youHow this skill is triggered — by the user, by Claude, or both
Slash command
/software-of-you:health-monitorThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Self-healing health checks for Software of You. Validates infrastructure, auto-repairs what it can, and generates a visual health dashboard.
Monitors PostgreSQL, MySQL, MongoDB health via CLI queries for connections, throughput, disk usage, cache ratios, locks, with alerting thresholds and automation guidance.
Sets up Supabase observability with dashboard reports, CLI inspect commands for Postgres diagnostics, pg_stat_statements, log drains, and alerting.
Diagnoses PostHog SDK integrations: identifies outdated SDKs, provides upgrade recommendations, and matches the SDK Health UI. Use when a user asks about SDK versions or health.
Share bugs, ideas, or general feedback.
Self-healing health checks for Software of You. Validates infrastructure, auto-repairs what it can, and generates a visual health dashboard.
/patrol — run all checks, auto-fix issues, generate health dashboard/health — quick status check (read-only, no repairs)Run the health check script:
# Full patrol with auto-repair + dashboard
python3 "${CLAUDE_PLUGIN_ROOT:-$(pwd)}/scripts/health_check.py" --fix --dashboard
# Read-only check
python3 "${CLAUDE_PLUGIN_ROOT:-$(pwd)}/scripts/health_check.py"
# JSON output for programmatic use
python3 "${CLAUDE_PLUGIN_ROOT:-$(pwd)}/scripts/health_check.py" --fix --json
| Check | What | Auto-Fix |
|---|---|---|
| Database integrity | All 31 tables + 9 views exist, SQLite integrity_check passes, no FK violations | Runs bootstrap to recreate missing tables/views |
| OAuth tokens | Token files exist, not expired, have refresh tokens | Auto-refreshes expired tokens |
| HTML views | Core views (dashboard, contacts, week-view, email-hub, nudges, timeline) exist with content | Flags missing views in soy_meta for regeneration |
| Backups | Backup directory has files, latest is < 48h old | Creates fresh backup |
| Sync freshness | Gmail, Calendar, Transcripts synced within 60m | Reports staleness (sync via pipeline) |
If the patrol flags HTML views for regeneration, regenerate them:
SELECT value FROM soy_meta WHERE key = 'views_need_regeneration';
If this returns a list, generate those views using the dashboard-generation skill.
Before any destructive database operation (wipe, reset, drop), always call:
from scripts.health_check import enforce_backup_before_destructive
enforce_backup_before_destructive("database_wipe")
Or via bash:
python3 -c "from scripts.health_check import enforce_backup_before_destructive; enforce_backup_before_destructive('your_operation')"
The patrol generates output/health.html showing: