Integration Health Monitor
Detect broken feeds and mismatches before they hit the close
A monitoring workflow that checks data pipelines between banks, PSPs, ERP systems, and spreadsheets, flags anomalies early, and recommends remediation steps so you are not scrambling on day one of close.
INGREDIENTS
PROMPT
Create a skill called "Integration Health Monitor". Every day: - Check freshness and completeness for each configured integration or feed - Detect duplicates, missing fields, and volume anomalies - Run control-total checks where applicable - Alert the responsible owner with a short remediation playbook - Keep an incident log and weekly trend summary Ask me to define: - The list of critical integrations - Expected update cadence for each - Alert thresholds and owners
How It Works
When integrations break, teams fall back to manual exports and spreadsheet stitching.
This recipe provides a daily integration heartbeat and early-warning system.
What You Get
- Daily integration status report with green, yellow, and red states
- Alert list with root-cause hypotheses and remediation steps
- Weekly trend report of recurring failures and manual-work drivers
Setup Steps
- List critical integrations and feeds
- Define expected refresh frequency for each
- Add health rules for freshness, duplicates, and balance checks
- Assign owners and escalation contacts
- Publish daily status and incident summaries
Tips
- Include subledger-to-GL control checks where possible
- Log recurring failures so system hardening becomes visible
- Alert with next actions, not just raw errors