vome-health: your Home Assistant's health, beside Claude Code

A side pane with Vome's health score for your Home Assistant, out of 100, and everything its check found, kept beside the conversation while Claude works through it.

• The score, big, in the terminal, with a heart beating beside it, coloured by how the home is doing, and when it was checked.
• What the check found, by severity: To fix, Worth doing, Fine, each with what to do about it. Devices flooding the recorder, entities left behind by removed integrations, automations that are off or never run, batteries not reporting, error noise.
• What Claude is working on: when Claude changes something a finding is about (one of its entities, the recorder for a flooding sensor, an automation for the automation findings), the finding is marked ✎ until a re-check settles it.
• The re-check: press r (or ask Claude). The new score rolls in, findings it no longer has are struck through ✓, and a better score gets fireworks.

The pane never fixes anything itself: Claude does, with its own tool calls, which you approve as usual. r only asks Vome for a fresh check.

Quickstart

1. Add the Vome app to Home Assistant, then get a key from it. No sign-up needed.

The button opens your own Home Assistant and adds the app's repository. Pick Vome and Install, open it, and on the Agent tab create a key.

2. Paste these into Claude Code, in a terminal (version 2.1.275 or newer):

/plugin install vome-connect --marketplace Vortitron/home-assistant-mcp

It asks for the key from step 1. Then:

/plugin install vome-health --marketplace Vortitron/home-assistant-mcp
/reload-plugins

Or all four Vome panes at once (automations, ESPHome, health and a working dashboard): /plugin install vome-panes --marketplace Vortitron/home-assistant-mcp.

3. Ask Claude "How healthy is my Home Assistant? Fix what you can." The pane opens with the score. No score yet? Claude runs the first check, which takes a couple of minutes.

Allow its reads (auto mode)

The pane reads the report in the background while a check runs, with the MCP's read-only vome_health_report; r calls vome_health_check. In auto mode, Claude Code refuses background calls that no request of yours is behind unless they're allowed by name, and the pane then shows the exact line. With vome-connect:

{ "permissions": { "allow": ["mcp__plugin_vome-connect_vome__vome_health_report"] } }

Keys

| | | | --- | --- | | /health | open the pane and read the latest report | | r | run a fresh check | | d | each finding's evidence, not just its title | | b | the animation off or on (remembered) |

Privacy

The report is Vome's health check of your own Home Assistant, read from the sensor the Vome integration keeps it on (sensor.vome_health_score or similar). The pane reads it through the connection Claude Code already has; nothing is sent anywhere else.