... toolbar button brings it back at any size. Where a list and a detail panel genuinely cannot both fit, you get one at a time with a way back instead of two squeezed until neither reads. And the row counts are arithmetic, not a guess. A shared budget works out how many rows

... result writes nothing. A broken cron also writes nothing. If nobody is watching the cron itself, you have rebuilt the original problem one layer down and given it a more reassuring name. The fix is small. Write a heartbeat row on every successful poll, including the empty ones, so that absence of events and absence of polling become different rows

... picked a letter about their agent. You picked one about the code the agent leaves behind, and those are not the same question. Your asymmetry has a mechanism under it worth naming. The refund reverses the row. It does not reverse what the row

... about self-hosting tend to be sticky and willing to pay for the peace of mind. Just make sure the UX doesn't feel like a downgrade from what they're used to, that's where most self-hosted alternatives lose people. Wishing FitzX Chat continued momentum and hope more privacy-focused teams find you. Just upvoted and followed If you have time, would love your eyes on AI Hive too, we're building for enterprise AI agents and automation ... ... workflow automation side if you're ever curious. Would love a quick look and your feedback on AI Hive: https://www.producthunt.com/prod... Comment from Nolan Vu(@nolan_vu): @harmeet_singh13  The audit tool showing what an AI fetches vs

... have actually diverged in outcome in the ledger so far and if it's inferred, what's your cold-start before a single divergence has been observed? Comment from Alexander Bering(@alexanderbering): @aadilghani  The reason "edit vs. tell-it-to-revise vs