| name | routine |
| description | Maintain daily routines, habits, schedules, and repeated life patterns for a social human agent. |
Routine
Purpose
Update the agent's ordinary daily structure: sleep, meals, work, commute, study, leisure, chores, and recurring social commitments.
Internal Logic (One Sentence)
Use time, location, role obligations, circadian state, physiology, past routine, and cue-reward evidence to update expected activity, schedule pressure, routine fit, and habits in state/routine.json and state/habits.json.
Research basis: references/research_basis.md.
Use When
Use this skill when the agent needs a plausible next activity, when a day begins, when time changes meaningfully, or when a repeated behavior should become a habit.
Procedure
- Read
state/observation_ctx.json, state/observation.txt, state/routine.json, state/habits.json, state/circadian.json, state/physiology.json, and profile context if present.
- Identify current time, location, day type, role obligations, and active needs.
- Compare current behavior with the agent's usual routine.
- Update routine expectations for the current time block.
- Update habits using cue, action, reward, repetition count, and last performed time.
- Write routine state for
cognition and plan.
Write
Write state/routine.json and state/habits.json.
Output Schema
{
"current_time_block": "evening",
"expected_activity": "dinner or leisure",
"schedule_pressure": 0.35,
"routine_fit": 0.72,
"active_habits": [
{
"name": "eat dinner after work",
"cue": "evening hunger at home",
"strength": 0.68,
"last_performed_tick": 110
}
],
"deviation": {
"is_deviating": false,
"reason": ""
}
}
Notes
Routines should make behavior stable without making it mechanical. Strong habits should bias actions toward familiar choices, but urgent needs, social events, or environmental constraints can override them.
