Azure Pricing MCP has well-structured tool definitions with complete input schemas and reasonable descriptions. All 5 tools are explicitly registered with proper JSON Schema. However, descriptions lack LLM-optimization guidance (e.g., no 'when to use' context), parameter descriptions are generic, and output schemas are entirely undocumented. Error handling is absent, no guidance on retryability, user-fixable errors, or recovery paths. Tool composition is sound (each does one thing), but missing pagination guidance for large result sets and no documentation of what fields are returned.

• Document output schema for each tool. E.g., query_azure_prices returns {items: [{retailPrice, unitPrice, armRegionName, meterName, ...}], count, nextPageLink}. This enables LLMs to chain calls and extract IDs.
• Add 'when to use' guidance to tool descriptions. E.g., 'Use query_azure_prices for detailed filtering; use compare_vm_prices to compare SKUs across regions; use get_reservation_prices for reserved instance pricing.'
• Document error scenarios and recovery paths. E.g., 'If API returns 429 (rate limit), retry after 60 seconds. If filter is invalid, simplify the customFilter expression or use named parameters instead.'
• Provide OData filter examples in customFilter description. E.g., 'contains(meterName, "Spot"), startswith(serviceName, "Virtual"), isPrimaryMeterRegion eq true'.
• Add pagination guidance: 'Results are limited to maxResults (default 100). If NextPageLink is present, use it to fetch the next page. Large queries may require multiple calls.'