auditing-warehouse-source-health

Run a source health audit that calls the data-health-issues endpoint to list all failed sources and sync schemas. The audit groups issues by severity, showing source-level errors first since they cascade to every schema underneath.

A source error means the entire external data connection is broken, so every schema under it stops syncing. A sync schema failure affects only one table, typically from schema drift like a renamed column or a billing limit being reached.

No, the data-health-issues endpoint only reports active source and sync failures. Webhook problems require calling webhook-info-retrieve per source to check whether the webhook exists, is enabled remotely, or has external errors.

A schema can report Completed status while its last_synced_at timestamp is far older than its sync frequency allows. Cross-check external-data-schemas-list for schemas whose last sync is stale relative to their configured frequency.

No, the audit is strictly read-only and never calls destructive tools. It reports issues and recommends hand-offs to diagnosis or tuning skills, and any fix requires explicit user confirmation before execution.

Use a full audit when you want a project-wide triage of what is broken. If you already know which sync is failing and want a deep-dive on that one failure, go directly to the diagnosing-failed-warehouse-syncs skill.