The installer wrote the same service list (ollama, homeassistant, mysql,
mariadb, nginx, apache2, docker) onto every host regardless of what it
actually runs, causing false "Service Down" alerts on hosts that never
installed those services. Default is now empty; watch_services should be
set per-host to match what's actually running there.
public_html/agent/ is what agents download for self-update.
It was 5 days out of date — missing the version-in-heartbeat fix
and all other v3.1 changes. Now mirrors agent/jarvis-agent.py exactly.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Add version column to registered_agents table
- Agents send version on registration (Linux 3.1, Windows 3.0, macOS 3.0)
- workers_list API returns latest_versions per platform
- Workers tab: VERSION column with green check (up-to-date) or red (outdated)
- Outdated agents highlight row and show blue UPDATE button
- Up-to-date agents show dimmed UPDATE button
- Update button dispatches update command immediately
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Previously agents only registered when api_key was missing (first run).
After updating to v3.0 with screenshot capability, restarted agents never
refreshed their capabilities in the DB. Now register() is called every
startup — server does UPDATE on existing agent_id so api_key is preserved.
self_update(cfg) and shell allow_shell_commands check both referenced cfg
from run() scope, but execute_command() is a standalone function. Fixed by
calling load_config() locally in each branch that needs it.
- 4-tier chat: HA control → Ollama → Groq → Claude
- Push-based agent system with heartbeat/metrics
- Network monitoring, alerts, Proxmox, Home Assistant
- Windows + Linux agent installers
- Stats cache cron, facts collector, KB engine
myron