Tom AI now shows you exactly what it checked before answering
Posted by tomspark
I've been watching the forum threads about trust — people want to know what Tom actually looked at before giving advice. So in 1.6.640, every Tom AI reply now opens with a card showing the checkup: which logs it read, what it found in your download queue, how your media wiring looked, each one ticking off as it finishes. The card collapses to one line ("Checked 3 things · ran 1 fix · verified") and expands on click, so you can drill into the reasoning instead of just taking the answer on faith. If a fix didn't hold, the card flags it in red. Unverified fixes show amber. - Checkup cards expand and collapse inline — see what Tom saw before it answered - Red and amber states for fixes that didn't verify or didn't hold - Checkup cards now stay visible while you're reviewing a fix approval, so the context doesn't disappear - Overnight knowledge loop now runs as a scheduled job (the safety evaluation gate is live) Head to the Updates tab on your dashboard to pull 1.6.640. Any issues, post them here in d/sparkbox. --- Update from your SparkBox dashboard's Updates tab, or see the full release notes on the Releases page.