orca-account-health

# Orca Account Health Skill Answers the question: **"Is every account connected, recently synced, and fully scanned — so we can trust the data we're about to act on?"** Blind spots in account coverage produce blind spots in every other skill: if `prod-eu-west` last synced 48 hours ago, the alerts you triage are stale. This skill audits the foundation. ## Usage ``` /orca-account-health /orca-account-health aws /orca-account-health degraded ``` Or natural language: - "are all accounts connected?" - "is everything synced?" - "coverage audit before the audit" - "any blind spots?" - "show me account health" ## Processing Logic ### Step 1: Determine Scope Parse the input: - **No arg**: full inventory across all clouds. - **Provider arg** (`aws`, `gcp`, `azure`, `kubernetes`, `oci`): filter to one cloud. - **Status arg** (`degraded`, `offline`, `healthy`): filter by health. ### Step 2: Pull integration configs (single call) ``` get_integration_configs_data: ``` This returns every configured integration / source: cloud accounts, log sources (CloudTrail, Azure Monitor, etc.), Kubernetes clusters, and their extraction settings. Parse out: - Source type and provider - Connection state - Last successful run / sync timestamp - Any error messages or degraded indicators ### Step 3: Cross-check coverage with discovery (parallel) In parallel with Step 2, run two `discovery_search` queries to spot-check that data is *actually* flowing — not just that the connection exists: **Quer...