Two tools with basic schemas and minimal descriptions. Both tools have input schemas with proper types (number, string enum), but descriptions are extremely short (10-40 chars) and lack context for LLM selection. No output schema documentation. Error handling present only for divide-by-zero in 'calculate'. Tool names are action verbs but lack specificity, 'add' and 'calculate' don't distinguish use cases. Parameters have types but descriptions are absent or trivial. No guidance on when to use 'add' vs 'calculate'. Schemas are visible and valid, preventing a lower score, but quality gaps are substantial.

• Expand tool descriptions to 50-150 chars. Example: 'add' → 'Add two numbers and return the sum. Use this for simple addition; use calculate() for subtract, multiply, or divide operations.'
• Add parameter descriptions for all inputs. Example for 'a' and 'b': 'A numeric value (integer or float). No range limits; precision depends on JavaScript number type.'
• Document output schema explicitly. Example: 'Returns an object with {result: number} on success, or {error: string} on failure (e.g., divide by zero).'
• Rename tools for clarity: 'add' → 'add_numbers', 'calculate' → 'calculate_arithmetic'. This helps LLMs distinguish when each is appropriate.
• Add input validation and error messages. Catch non-numeric inputs and return: 'Invalid input: a and b must be numbers. Got a={type(a)}, b={type(b)}.' This enables LLM self-correction.