Three tools with basic schemas and descriptions, but significant gaps in parameter documentation, output schema clarity, and error handling guidance. list_images has no input parameters documented. create_image and create_vm lack descriptions for several parameters (e.g., image_type enum values, memory constraints). No output schemas documented for any tool. Error handling returns generic error dicts without recovery guidance. Tool names follow verb_noun convention (positive), but descriptions are minimal (avg ~50 chars, below 194-char baseline). No tool annotations (readOnlyHint, destructiveHint) despite clear risk levels.

• Document output schemas for all tools. E.g., list_images should return {images: [{name, size_bytes, type, image_id}], total_count, next_cursor}. create_image/create_vm should return {task_id, resource_id, status, created_at}.
• Add enum constraints to image_type parameter: 'Type of image (DISK_IMAGE, ISO_IMAGE, or VMDK_IMAGE)' instead of generic 'etc.'
• Expand create_vm memory_size_gb description: 'Memory size in gigabytes (minimum 0.064 GB = 64MB, maximum 12288 GB = 12TB)' with explicit bounds.
• Implement pagination in list_images: add offset/limit params, return total_count, and document 'Call repeatedly with offset+=50 to fetch all images'.
• Add tool annotations via FastMCP: @mcp.tool(annotations={'readOnlyHint': True}) for list_images; @mcp.tool(annotations={'destructiveHint': True}) for create_image/create_vm.