Server has 13 tools with basic schemas and descriptions, but significant quality gaps. Most tool descriptions are 1-2 sentences (avg ~80 chars, below the 194-char baseline). Parameter descriptions are minimal or missing context. No output schemas documented. Error handling returns raw strings without recovery guidance. Tools lack annotations (readOnlyHint, destructiveHint). Critical security issue: k8sgpt_configure_auth accepts api_key as a parameter, violating secret-injection pattern. Weather tools (get_alerts, get_forecast) are unrelated to AKS domain, suggesting scope creep. No pagination support despite tools that could return large result sets.

• Move api_key from k8sgpt_configure_auth parameter to server-side environment variable (K8SGPT_API_KEY). Update tool to read from env and return success/failure only.
• Document output schemas for all tools. E.g., get_aks_clusters returns [{name, resourceGroup, location, kubernetesVersion, provisioningState}]. Include field types and whether fields are always present.
• Add dry-run parameter to create_aks_cluster and delete_aks_cluster. Return what would happen without executing. Require explicit confirmation for destructive operations.
• Enhance error messages with recovery guidance. E.g., 'Error: Resource group not found. Try get_aks_clusters() to list available groups' or 'Retryable: Azure CLI timeout. Retry in 5 seconds.'
• Add tool annotations to FastMCP tool registration: @mcp.tool(annotations={'readOnlyHint': True}) for read tools, @mcp.tool(annotations={'destructiveHint': True}) for delete/create tools.

Inferred effective spec: <=2025-11-25.

• Relies on Logging (deprecated) - log to stderr or use OpenTelemetry