Two tools with minimal definitions. Both have basic schemas and descriptions, but lack depth, parameter documentation, and error guidance. 'calculator' description is 60 chars (acceptable baseline ~194), but 'search' is only 54 chars. Neither tool has parameter descriptions beyond the bare type. No output schema documentation. No error handling guidance. No tool annotations. Naming is acceptable (verb_noun pattern), but overall quality is well below production baseline.

• Add parameter descriptions to both tools. For 'calculator': 'Arithmetic expression using operators +, -, *, /, %, ** (e.g., "12*8" or "(3+4)*2"). Must be a valid Python expression.' For 'search': 'Search keyword or phrase to match against knowledge base entries. Case-insensitive. Returns up to 10 matching results.'
• Document output schema. Specify that calculator returns a numeric result as a string, and search returns a list of matching entries with fields like 'title', 'content', 'relevance_score'.
• Add error handling guidance. For calculator: 'If expression is invalid (e.g., syntax error or division by zero), the tool returns an error message. Verify the expression syntax and retry.' For search: 'If no results found, try a shorter or more general keyword.'
• Expand tool descriptions to 100-150 chars. Include WHEN to use: 'Use calculator for arithmetic evaluation when you need to compute numeric results. Supports basic math operations.' For search: 'Use search to find information in the knowledge base. Call this first to discover available topics before answering user questions.'
• Add tool annotations to schema. Include 'readOnlyHint': true for both tools to signal they are safe to call without side effects. This enables LLMs to retry safely on transient failures.