Single tool with basic schema and Korean-language description. Tool name 'generate-mock-data' follows verb_noun convention but lacks clarity on output format. Input schema is present with proper types (number, array) and constraints (min/max), but parameter descriptions are minimal and lack format guidance. No output schema documented. Error handling validates Zod schema but provides generic error messages. Tool is read-only and low-risk, but definition quality falls short of production standards.

• Translate tool description to English and expand to 50-150 characters: 'Generate mock data records with custom fields (name, email, phone, address, company, job, city, country). Useful for testing and prototyping.'
• Document the output schema explicitly: 'Returns an array of objects, each with the requested fields populated with realistic fake data. Example: [{"name": "John Doe", "email": "[email protected]"}]'
• Replace 'fields' description examples with an enum constraint in the schema or a comprehensive list: 'Supported fields: name, email, phone, address, company, job, city, country. Unsupported fields default to random strings.'
• Enhance error messages with actionable recovery: catch unsupported field names and return 'Field "ssn" not supported. Try one of: name, email, phone, address, company, job, city, country.'
• Add parameter descriptions in English: 'count: Number of mock records to generate (1-100, default 1)' and 'fields: Array of field names to include in each record (required, min 1 field).'