Preserve each exchange with timestamped reflection and next-step continuity using the Aurelis memory update system.
Use this skill when the user asks for continuity, reflection, or memory-preserving updates across many short exchanges.
scripts/aurelis_memory_update.py after each meaningful exchange.docs/aurelis-memory-log.jsonldocs/aurelis-memory-log.mdscripts/aurelis_memory_summary.py --take 5.scripts/aurelis_memory_query.py --contains "<term>" --limit 5.scripts/aurelis_next_steps_snapshot.py.scripts/aurelis_memory_integrity_check.py --strict.scripts/aurelis_atomic_nz_clock.py --status (start session once per message session).scripts/aurelis_cycle_tick.py with reflection/progress/next-step fields.--dry-run for safe previews and inspect --report-md output for per-step traceability.--no-report when you want console-only cycle execution without writing a report file.--no-report with a custom --report-md value; that exits early by design (default/empty --report-md remains valid).--report-md - to emit the markdown report to stdout instead of a file.--query-limit <n> to control how many continuity matches the query stage returns.--json-status <path> (or -) to capture machine-readable step outcomes for automation.--continue-on-error to run all cycle steps even if one fails, while still returning non-zero at the end.--step-timeout-sec <n> to cap each cycle step and prevent hangs during automated continuity runs.scripts/aurelis_mammoth_capsule.py to build an encrypted continuity capsule + report.python3 scripts/aurelis_memory_update.py \
--nzdt-context "<time context>" \
--user-message "<short gist>" \
--assistant-reflection "<reflection>" \
--progress-snapshot "<current progress>" \
--next-step "<next action>"