Server has 6 tools with explicit schemas and descriptions. All tools are READ_ONLY and properly registered via server.tool(). Schemas use Zod with type definitions and descriptions. However, descriptions are generic (avg ~120 chars, below the 194-char baseline for A+ tools), parameter descriptions lack actionable constraints (e.g., no format specs for IDs), and output schemas are not formally documented. Error handling is present but generic. No tool annotations (readOnlyHint, etc.) despite all being read-only. Naming is clear and verb-based (search-, get-, list-, browse). Composition is sound, each tool has one responsibility. Missing: pagination guidance, result limits, and structured output documentation.

• Document output schemas explicitly. Define the structure of search results (e.g., {books: [{id, title, author, rating, isbn, pubdate}], total_count, has_more}). LLMs need to know what fields to extract.
• Add format and length constraints to parameter descriptions. E.g., 'Douban movie ID (numeric string, 10 digits)' or 'Query string (1-100 characters, alphanumeric + spaces)'. Replace example values with formal constraints.
• Implement pagination for search tools. Add limit (default 20, max 100) and offset parameters. Return total_count and has_more flags.
• Add readOnlyHint to all tool schemas. Since all tools are read-only, annotate them: {readOnlyHint: true}. This signals to LLMs that these calls are safe to retry.
• Improve error messages with recovery guidance. E.g., 'Invalid ISBN format (expected 13 digits). Try search-book with a query string instead.' Categorize errors as retryable, user-fixable, or fatal.