Hi everyone,
I have been experimenting with a small local tool for reviewing Home Assistant configuration exports and would appreciate feedback from people who run more varied setups than mine.
The project is called Home Assistant Audit Lite:
The goal is intentionally limited:
- run locally,
- read only an exported/redacted Home Assistant configuration folder,
- do not connect to Home Assistant,
- do not call the network,
- do not modify any files,
- avoid printing secret values,
- produce a Markdown or JSON report with review prompts.
It currently checks things like:
- presence of common config files,
- secret-like key names without printing values,
- references to entities in automations/scripts/config,
- optionally redacted log files for WARNING/ERROR/Traceback patterns,
- basic Energy dashboard sanity checks around
device_class,state_class, units like W/kW/Wh/kWh/mÂł, - recorder configuration hints,
- MQTT / Zigbee2MQTT sanity hints,
- backup/update readiness review prompts.
Example usage:
python3 ha-audit-lite-v0.2.py /path/to/ha-export /path/to/report.md
JSON output is also supported:
python3 ha-audit-lite-v0.2.py /path/to/ha-export /path/to/report.json --format json
The findings are meant as review prompts, not as proof that a production setup is broken. The tool uses conservative line/regex analysis rather than trying to fully understand every Home Assistant integration or template.
I am mainly looking for feedback on:
- Are there checks that would be useful for real Home Assistant maintenance but are still safe to run locally?
- Are any current checks likely to create too much noise or misleading output?
- Are there common configuration patterns this kind of tool should avoid flagging?
- Is the “local, read-only, no upload” boundary clear enough?
I am especially interested in feedback from people with Energy dashboards, MQTT/Zigbee2MQTT setups, larger automation files, or package-based configurations.
Thanks.