Account Monitoring

Checks Meta ad account health — policy status, billing flags, API error rates, and unusual spend patterns outside normal campaign bounds.

Chuck3PointZero Updated

File contents

Skill: Account Monitoring

Used exclusively by account-sentinel. Fetches account-level signals that sit above the campaign layer — policy status, billing health, and aggregate spend anomalies that no single campaign metric would surface.

Workflow

  1. Call meta-ads.get_account_status — returns policy status, billing flags, and any active account restrictions.
  2. Call meta-ads.get_campaign_insights with a short lookback window — compute aggregate spend vs. historical baseline.
  3. If aggregate spend deviates by more than 2× the 7-day average: flag as unusual-spend-pattern-detected.
  4. If any policy restriction is ACTIVE or PENDING_DELETION: flag as account-policy-violation-detected.
  5. If the meta-ads tool returns repeated 4xx or 5xx codes across the session: flag as api-error-rate-above-threshold.
  6. Write findings to memory/account-health-{date}.json.
  7. Return: list of active flags (empty list = clean).

All non-empty flag lists are escalated by account-sentinel without autonomous action.

Chuck3PointZero/OAA/tree/main/examples/performance-marketing/skills/account-monitoring commit d107c383d5

Frequently asked questions

npx skillmds@latest add chuck3pointzero/account-monitoring