Two tools with basic schemas and minimal descriptions. Both tools have input schemas with proper types (number, string, enum), but descriptions are extremely short (10-50 chars) and lack context for LLM selection. No output schema documentation. Error handling is minimal (divide-by-zero check in 'calculate' returns text, not structured error). No parameter descriptions beyond type hints. Tool names are generic verbs without clear domain context. Schemas are present but descriptions fall well below the 50-200 char baseline for A+ tools.

• Expand tool descriptions to 50-200 chars. Example for 'add': 'Adds two numbers and returns their sum. Use this for simple two-operand addition. For complex expressions with multiple operations, use calculate() instead.'
• Add descriptions to all parameters. Example: 'a: First operand (number). b: Second operand (number).' For 'calculate', add: 'operation: Arithmetic operation to perform (add, subtract, multiply, divide). a: First operand. b: Second operand.'
• Document output schema explicitly. Both tools return {type: 'text', text: string}. Declare this in tool metadata or response documentation.
• Implement structured error responses. Replace 'Error: Cannot divide by zero' with {error: 'DIVISION_BY_ZERO', message: 'Cannot divide by zero. Ensure b is non-zero.', retryable: false}.
• Rename tools for clarity. Consider 'compute_sum' instead of 'add', and 'compute_arithmetic' instead of 'calculate' to better signal intent and distinguish use cases.