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).
Complements skill-audit-weekly (frontmatter/credentials lint). This task
answers a different question: of the ~120 skills we have, which ones never
trigger? The failure mode caught here is Atlas/GBrain (commit 62b0192) —
a skill existed, looked healthy, but silently never matched real user
phrasing.
Runs Monday 09:00 SAST. Writes ~/workspace/reports/skill-audit-YYYY-MM-DD.html
(surfaces on the MC dashboard under /reports). Telegrams Elmar only when a
previously-active skill has gone silent or a near-miss recurs 3+ times.
python3 /home/lucienne/workspace/scripts/skill_audit.py
Format: minute hour day month weekday
| Started | Status | Duration | Log |
|---|---|---|---|
| 5d ago | completed | 35s | 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.