Critical gaps across all dimensions. Tool definitions are inferred from file references rather than explicitly visible in source code. No input schemas are visible in the provided code. Descriptions are minimal (6-25 chars), far below the 50-200 char baseline for LLM-optimized tools. No parameter documentation. No output schemas documented. The server.py file is referenced but not provided, making it impossible to verify actual tool registration, schema definitions, or parameter handling. Based on visible evidence only: tools appear to be defined but schemas and detailed descriptions are not shown.

• Provide complete server.py source code showing actual tool registration, input schemas, and parameter definitions. Current evaluation is based on inferred tool existence only.
• Expand all tool descriptions to 50-200 characters. Example: 'get_models' should be 'Retrieve all available Looker models. Use this to discover what data sources are available before querying. Returns model names and descriptions.'
• Add explicit input schemas with parameter types, descriptions, and constraints. Example for 'query': {"model": {"type": "string", "description": "Looker model name (e.g., 'ecommerce')"}, "explore": {"type": "string", "description": "Explore name within the model"}, "dimensions": {"type": "array", "items": {"type": "string"}, "description": "Dimension names to include in results"}}
• Document output schemas for all tools. Example for 'get_models': {"type": "object", "properties": {"models": {"type": "array", "items": {"type": "object", "properties": {"name": {"type": "string"}, "description": {"type": "string"}, "project_name": {"type": "string"}}}}, "total_count": {"type": "integer"}}}
• Add error handling with recovery guidance. Example: 'If query fails with "Model not found", call get_models() first to see available models. If query times out, reduce dimensions or add filters to limit result size.'