What problem does it solve? Teams integrating Azure's public pricing endpoint ship callers that silently undercount SKUs, miss savings-plan rates, or reconcile reference-only currency prices against invoices. This Skill encodes the full API contract and operational discipline so pricing callers return defensible, reproducible results. ## Core Features & Use Cases - API Contract Reference: Endpoint, both API versions, case-sensitive $filter rules, the small OData operator surface, priceType and 26-value serviceFamily enums, and the complete response schema. - Pagination & Hardening Patterns: Walk NextPageLink to null (1,000 records/page), exponential backoff on 429/5xx, and 24-hour cache TTL keyed on filter, currency, and API version. - Anti-Pattern Detection: Flags production traps like lowercase filter values returning empty results, reservation retailPrice treated as hourly, and mixing Consumption with DevTestConsumption rows. - Use Case: Build a monthly cron that snapshots Compute prices per region, diffs them to detect Azure price drift, and feeds a FinOps budget model — without ever mistaking catalog prices for invoiced billing. ## Quick Start Ask the AI to write a Python caller that fetches all eastus Consumption prices for Standard_D4s_v5 from the Azure Retail Prices API with full pagination and backoff.
