Hass Bench - dry-run the Home Assistant automation before you reload it

Paste the Home Assistant automation you just wrote and find out what it will actually do before you reload it. The page reads the YAML in your browser with a reader written against Home Assistant's own PyYAML-derived loader - so it knows that an unquoted `to: on` is the boolean true, which is why that trigger silently never matches a switch's string state, and that an unquoted `at: 7:30:00` is the integer 27000. Forty deterministic checks run over it free, then a real dry-run simulator: set the state of every entity the automation names, plus the time, the sun and the weekday, and watch the trace - which trigger fires, which condition blocks, which actions run, with the clock advancing through every delay and each action changing the simulated world as it goes. Templates are evaluated by a Jinja-subset parser with no eval anywhere, and anything it cannot evaluate is listed rather than assumed true. Then four lanes over the same document: review it trigger by trigger, explain it in plain English for whoever did not write it, harden it into the current schema with every YAML 1.1 trap quoted, or generate the Lovelace card that watches it. Every finding the browser made is sent as a fact the model must address, and the page shows you which ones it skipped; the hardened rewrite is re-read by the same reader before you see it. Derived from four agent skills and credited to all of them: @bradsjm/home-assistant-automation-scripts (automations, scripts and blueprints - triggers, conditions, actions, run modes and troubleshooting), @homeassistant-ai/home-assistant-best-practices, @claude-office-skills/home-assistant-automation and @bradsjm/home-assistant-dashboards-cards (Lovelace views, cards and resources). An independent derived work, not affiliated with those skills' authors, with the Home Assistant project, with Nabu Casa, or with any integration it names.

Every public app is built from a security-scanned skill and must pass a clean scan — skill and frontend — before it can be listed. Have a skill of your own? Turn it into an app — or read the step-by-step walkthrough.