Example project — Frond is not real.
Authored demo content for a fictional plant-care app, rendered by the same parsers and held to the same invariants as a real project's dashboard. This is the system in use, three months in.
Future considerations3
Known directions parked until a trigger fires. The unit here is the trigger, not the task — items promote out (to the punch list, a phase board, or feature scope) when their trigger fires, and leave when a phase ships them.
Waiting on: a support request from two users watering the same plant, or the first paid tier discussion.
Two people, one plant list. Needs a real permission model and turns the digest-time question (Open Questions §3) into a genuine conflict. Big enough to be its own milestone, not a phase.
Effort: large
Waiting on: the first user outside the northern hemisphere, or any marketing outside it.
Seasonal adjustment is hardcoded to one hemisphere (ROADMAP.md → Key Considerations). The fix is small and the embarrassment is not, so this graduates the moment it is real.
Effort: small
Waiting on: watering retention holds for two months, proving the habit exists to attach to.
Explicitly out of the first milestone. Only worth building once watering is a habit, because it rides on the same digest and would otherwise dilute it.
Effort: medium