Two simple mathematical tools with basic descriptions and schemas. Both tools have proper type definitions (integer inputs) and parameter descriptions. However, descriptions are minimal (under 50 chars), lacking context on when to use each tool or what constitutes valid input ranges. No output schema documentation. No error handling guidance for edge cases (e.g., negative factorials, large Fibonacci indices). Tools lack verb-noun naming convention (fibonacci/factorial are nouns, not verb_noun). No parameter constraints (min/max bounds). Logging is present but tools do not leverage structured error responses or recovery guidance.

• Rename tools to verb_noun convention: 'calculate_fibonacci' and 'calculate_factorial'. This signals intent and helps LLMs disambiguate when many tools are available.
• Expand descriptions to 100-150 chars. Include valid input range (e.g., 'Calculate the nth Fibonacci number. n must be 0-100. Returns the nth number in the sequence.'), when to use it, and what it returns.
• Add parameter constraints: 'n (int, 0-100): The position in the Fibonacci sequence. Must be non-negative.' This prevents LLMs from passing invalid values.
• Document output schema in docstring: 'Returns: int, The computed Fibonacci or factorial value.' This enables LLMs to plan downstream operations.
• Enhance error messages with recovery guidance: 'Factorial is not defined for negative numbers. n must be >= 0. Try a non-negative value.' This guides LLM self-correction.

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

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