OVERRIDES runtime profile: uses direct_python (plain Python, no model) because the command chain never invokes the claude CLI or any LLM API — pure-infra task; scheduler provider env injection is a no-op (MC-4942 U12 sweep).
First of each month, 07:00 SAST. Read-only structural audit of Luci's
auto-memory store (~/.claude/projects/-home-lucienne-workspace/memory/).
Detects: orphan files (not in MEMORY.md index), dead index links, dead wikilinks, invalid frontmatter types, stale snapshot memories (>90d).
memory_audit.py exits 0 when clean and non-zero (issue-category count)
when it finds problems — the || branch fires a Telegram report only on
findings. Clean runs are silent. The audit does NOT auto-fix; Luci triages
the report (conservative-retirement rule — present, don't cut).
runtime_profile: claude_glm is nominal — the command is pure Python with
no claude/API invocation, matching the sibling memory-backup task.
python3 /home/lucienne/workspace/scripts/memory_audit.py > /tmp/memaudit.txt 2>&1 || python3 /home/lucienne/workspace/scripts/notify.py "Memory hygiene audit found issues — triage needed:" "$(cat /tmp/memaudit.txt)"
Format: minute hour day month weekday
| Started | Status | Duration | Log |
|---|---|---|---|
| 4d ago | completed | 0s | View log |
Mission Control is the board for your delegated work: requests come in, Luci coordinates the next step, and evidence stays visible for review.
Luci is your always-on assistant for routing, status updates, and follow-through. Operators can still open deeper evidence when needed.