system note

QuestOps: A Daily Starting Point

Open the Pi agent. A blank prompt stares back. No urgent bugs, no deadline pressure, just the cursor blinking, waiting for a decision. That moment costs more than most people admit.

QuestOps skips the friction. It reads active projects and surfaces five daily quests, one from each category: Build, Research, Content, Distribution, Kill. Build means a feature or config to ship. Research is a question worth investigating. Content is something to write, record, or publish. Distribution pushes it to a channel that matters.

Kill identifies something worth stopping: a feature that pulls attention, a process that no longer earns its place. Pick one quest, ship it, attach proof, score points. The loop runs entirely inside Pi. No dashboard, no signup, no separate app.

The categories are not arbitrary. They map to how work actually breaks down when every day stops being treated as a new emergency. It is a starting point, not a solution. The blank screen has not felt blank since.

Author's note — May 29, 2026

After a first day of use, the core loop is staying. QuestOps did what it was supposed to do: it made the next action obvious, surfaced a stale goal, and gave the Pi session a starting point without adding another app.

The research pass compared general gamified productivity tools like Habitica and Todoist Karma with developer-native experiments like Questly, CodeQuest, and GitHub-style achievements. The useful gap is not "make work cute." It is a terminal-native quest layer that uses real project signals and avoids fake motivation.

So the decision is conservative: keep Slice α as-is for now. No punishment mechanics, no opaque scoring redesign, no full XP system yet. Points stay because they create a useful scoreboard, but the scoring model needs more live use before it deserves more complexity.

Related: agent-house rooms · field notes · practical AI systems